From fredericroux at HOTMAIL.DE Fri Jan 4 11:04:03 2008 From: fredericroux at HOTMAIL.DE (Frederic Roux) Date: Fri, 4 Jan 2008 11:04:03 +0100 Subject: topoplot error Message-ID: Hi everybody, I've been trying to use the topoplotTFR function but constantly receive the error message 'Odd number of arguments?'. Has anyone ever encountered that problem? I used the following commands: cfg = []; cfg.layout = prepare_layout(data); cfg.baselinetype = 'relative'; cfg.xparam = 'time'; cfg.yparam = 'freq'; cfg.zparam = 'powspctrm'; cfg.xlim = 'maxmin'; cfg.ylim = 'maxmin'; cfg.zlim = 'maxmin'; topoplotTFR(cfg,data) topoplotTFR calls topoplotER which again calls topoplot. The problem seems to be related to the following lines of code: in topoplotER @ line 364: topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); and in topoplot @ line 957: if nargs> 2 if ~(round(nargs/2) == nargs/2) error('Odd number of input arguments??') end I don't get any further. Can anyone help me, please? Regards, Frederic _________________________________________________________________ Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! http://messenger.live.de/community/neuekontakte_adressimport.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From v.litvak at ION.UCL.AC.UK Fri Jan 4 11:24:35 2008 From: v.litvak at ION.UCL.AC.UK (Vladimir Litvak) Date: Fri, 4 Jan 2008 10:24:35 -0000 Subject: topoplot error In-Reply-To: Message-ID: This probably has to do with EEGLAB being above Fieldtrip in your Matlab path with EEGLAB's version of topoplot being called instead of Fieldtrip's. You need to put Fieldtrip on top and make sure it stays there because EEGLAB moves itself to the top each time it's called. Perhaps you should add a line modifying the path to your script. Best, Vladimir -----Original Message----- From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf Of Frederic Roux Sent: Friday, January 04, 2008 10:04 AM To: FIELDTRIP at NIC.SURFNET.NL Subject: [FIELDTRIP] topoplot error Hi everybody, I've been trying to use the topoplotTFR function but constantly receive the error message 'Odd number of arguments?'. Has anyone ever encountered that problem? I used the following commands: cfg = []; cfg.layout = prepare_layout(data); cfg.baselinetype = 'relative'; cfg.xparam = 'time'; cfg.yparam = 'freq'; cfg.zparam = 'powspctrm'; cfg.xlim = 'maxmin'; cfg.ylim = 'maxmin'; cfg.zlim = 'maxmin'; topoplotTFR(cfg,data) topoplotTFR calls topoplotER which again calls topoplot. The problem seems to be related to the following lines of code: in topoplotER @ line 364: topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); and in topoplot @ line 957: if nargs> 2 if ~(round(nargs/2) == nargs/2) error('Odd number of input arguments??') end I don't get any further. Can anyone help me, please? Regards, Frederic _________________________________________________________________ Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! http://messenger.live.de/community/neuekontakte_adressimport.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From fredericroux at HOTMAIL.DE Fri Jan 4 14:30:00 2008 From: fredericroux at HOTMAIL.DE (Frederic Roux) Date: Fri, 4 Jan 2008 14:30:00 +0100 Subject: topoplot error In-Reply-To: <001501c84ebc$071a4740$f0463ec1@sobell.ion.ucl.ac.uk> Message-ID: Hello Vladimir, thank you for the advice. That did it indeed. You saved my day :) ---------------------------------------- > Date: Fri, 4 Jan 2008 10:24:35 +0000 > From: v.litvak at ION.UCL.AC.UK > Subject: Re: [FIELDTRIP] topoplot error > To: FIELDTRIP at NIC.SURFNET.NL > > This probably has to do with EEGLAB being above Fieldtrip in your Matlab > path with EEGLAB's version of topoplot being called instead of Fieldtrip's. > You need to put Fieldtrip on top and make sure it stays there because EEGLAB > moves itself to the top each time it's called. Perhaps you should add a line > modifying the path to your script. > > Best, > > Vladimir > > -----Original Message----- > From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf > Of Frederic Roux > Sent: Friday, January 04, 2008 10:04 AM > To: FIELDTRIP at NIC.SURFNET.NL > Subject: [FIELDTRIP] topoplot error > > Hi everybody, > > I've been trying to use the topoplotTFR function but constantly receive the > error message 'Odd number of arguments?'. > Has anyone ever encountered that problem? > > I used the following commands: > > cfg = []; > cfg.layout = prepare_layout(data); > cfg.baselinetype = 'relative'; > cfg.xparam = 'time'; > cfg.yparam = 'freq'; > cfg.zparam = 'powspctrm'; > cfg.xlim = 'maxmin'; > cfg.ylim = 'maxmin'; > cfg.zlim = 'maxmin'; > > topoplotTFR(cfg,data) > > topoplotTFR calls topoplotER which again calls topoplot. > > The problem seems to be related to the following lines of code: > > in topoplotER @ line 364: > > topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); > > and in topoplot @ line 957: > > if nargs> 2 > if ~(round(nargs/2) == nargs/2) > error('Odd number of input arguments??') > end > > I don't get any further. Can anyone help me, please? > > Regards, > > Frederic > _________________________________________________________________ > Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! > http://messenger.live.de/community/neuekontakte_adressimport.html > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. _________________________________________________________________ Windows Live Fotogalerie: So einfach organisieren Sie Ihre Fotos! http://get.live.com/photogallery/overview ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Sun Jan 6 18:40:59 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Sun, 6 Jan 2008 17:40:59 +0000 Subject: topoplot error In-Reply-To: Message-ID: Hi everyone, Does anyone know where to find the standardised montreal " average" mri in ctf format? I see that there are some average mri's in the spm toolbox(spm2\canonical) but these are .mnc files in spm coordinates. I have two subjects without mris and have found that the localisation is much more accurate if i use another subjects ctf mri than if i use the mri's provided with spm (to produce the headmodel and for source interpolation). The MEG scanner used was a ctf machine so i thought that using a ctf mri average would give the best result. I have been careful to ensure that the coordinate system used both for the headmodel formation and for interpolation has been defined appropriately for the type of mri being used (eg cfg.coordinates = 'ctf'; or ='spm'). Any thoughts would be much appreciated. Also i cant seem to find the function beamformer_pcc.m which is called when using sourceanalysis.m, cfg. method ='pcc'; knowledge of its whereabouts would be greatly received! best wishes tom gilbertson ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at FCDONDERS.RU.NL Mon Jan 7 15:03:55 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Mon, 7 Jan 2008 15:03:55 +0100 Subject: automatic update wiki/ftp disabled for a few days Message-ID: Dear all The automatic updating of the documentation on the fieldtrip website and the automatic updating of the fieldtrip code on /home/common (internal to FC Donders) and on the external ftp server are TEMPORARY DISABLED. This is necessary due to migrating the cvs repository to a new server and to implementing new release management scripts. The automatic updating will be up again as soon as possible, but it might take a few days. best regards, Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Mon Jan 7 15:18:22 2008 From: jan.schoffelen at FCDONDERS.RU.NL (Jan Mathijs Schoffelen) Date: Mon, 7 Jan 2008 15:18:22 +0100 Subject: forward modelling without anatomical info Message-ID: Dear Tom, As far as I know, the 'ctf-format' standardised mri does not exist. Creating an anatomically accurate forward model, while the actual anatomical information is lacking, is tricky business. What kind of forward model do you actully use? The multiple sphere model based on the ctf-shape files? If so, you could try to come up with an 'averaged' multipsphere model (thus bypassing the need for a standardised MRI), although this is not trial. However, I wonder whether this would yield better results than just taking an 'averaged' single sphere model. I wonder what you mean by your statement that localisation is much more accurate when using ctf-formatted MRIs. Do you know where your activation should end up then? The function beamformer_pcc is as far as I know still in its experimental stage and only for use within the FC Donders Centre. However, beamformer_dics shares most of the functionality, but obviously not all of it. Regards, Jan-Mathijs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 7 16:59:09 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 7 Jan 2008 15:59:09 +0000 Subject: forward modelling without anatomical info In-Reply-To: <3c02ed1911ff6.478242be@ru.nl> Message-ID: dear jan mathijs, i have been generating a single shell model from the segmented mri. the localisation proceedure is for a tactile stimulus so when i see a contralateral S1 response i have some reason to belive that the algorithm is making some sense. I get the impression from your reply that there is no easy way to get around this lack of anatomical data and that i may just have to go and get the mri's done for this couple of subjects! thanks for your thoughts on this, tom. On 1/7/08, Jan Mathijs Schoffelen wrote: > > Dear Tom, > > As far as I know, the 'ctf-format' standardised mri does not exist. > Creating an anatomically accurate forward model, while the actual anatomical > information is lacking, is tricky business. What kind of forward model do > you actully use? The multiple sphere model based on the ctf-shape files? If > so, you could try to come up with an 'averaged' multipsphere model (thus > bypassing the need for a standardised MRI), although this is not > trial. However, I wonder whether this would yield better results than just > taking an 'averaged' single sphere model. > I wonder what you mean by your statement that localisation is much more > accurate when using ctf-formatted MRIs. Do you know where your activation > should end up then? > > The function beamformer_pcc is as far as I know still in its experimental > stage and only for use within the FC Donders Centre. However, > beamformer_dics shares most of the functionality, but obviously not all of > it. > > Regards, > > Jan-Mathijs > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From weisz at LYON.INSERM.FR Mon Jan 7 17:11:57 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Mon, 7 Jan 2008 17:11:57 +0100 Subject: forward modelling without anatomical info In-Reply-To: Message-ID: hi, on the fieldtrip FTP there is a standard BEM (standard_vol.matr). i think you could use the skull part as pseudo-headshape: vol.bnd(1).pnt this also fits with the standard MRI. there may be some transformation you have to apply to these "headshape" points so that the head fits nicely into the sensor array. then you can use prepare_localspheres and specify a lower number in cfg.spheremesh (perhaps something like 50; default is 4000). this may yield a crude but satisfactory solution if you are unable to get MRIs or individual headshapes. i'd be interested in the outcome. all the best, nathan On 07.01.2008, at 16:59, thomas gilbertson wrote: > dear jan mathijs, > > i have been generating a single shell model from the segmented mri. > the localisation proceedure is for a tactile stimulus so when i see > a contralateral S1 response i have some reason to belive that the > algorithm is making some sense. I get the impression from your reply > that there is no easy way to get around this lack of anatomical data > and that i may just have to go and get the mri's done for this > couple of subjects! > > thanks for your thoughts on this, > > tom. > > > On 1/7/08, Jan Mathijs Schoffelen > wrote: > Dear Tom, > > As far as I know, the 'ctf-format' standardised mri does not exist. > Creating an anatomically accurate forward model, while the actual > anatomical information is lacking, is tricky business. What kind of > forward model do you actully use? The multiple sphere model based on > the ctf-shape files? If so, you could try to come up with an > 'averaged' multipsphere model (thus bypassing the need for a > standardised MRI), although this is not trial. However, I wonder > whether this would yield better results than just taking an > 'averaged' single sphere model. > I wonder what you mean by your statement that localisation is much > more accurate when using ctf-formatted MRIs. Do you know where your > activation should end up then? > > The function beamformer_pcc is as far as I know still in its > experimental stage and only for use within the FC Donders Centre. > However, beamformer_dics shares most of the functionality, but > obviously not all of it. > > Regards, > > Jan-Mathijs > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 7 17:17:11 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 7 Jan 2008 16:17:11 +0000 Subject: forward modelling without anatomical info In-Reply-To: <99082A9C-EC3C-4EDE-884F-5D3D5AD3FADB@lyon.inserm.fr> Message-ID: thanks nathan, ill give it a try! tom On 1/7/08, Nathan Weisz wrote: > > hi, > > > on the fieldtrip FTP there is a standard BEM (standard_vol.matr). i think > you could use the skull part as pseudo-headshape: > vol.bnd(1).pnt > > > this also fits with the standard MRI. > > > there may be some transformation you have to apply to these "headshape" > points so that the head fits nicely into the sensor array. > then you can use prepare_localspheres and specify a lower number in > cfg.spheremesh (perhaps something like 50; default is 4000). > > > this may yield a crude but satisfactory solution if you are unable to get > MRIs or individual headshapes. > > > i'd be interested in the outcome. > > > all the best, > nathan > > > > > > > > On 07.01.2008, at 16:59, thomas gilbertson wrote: > > dear jan mathijs, > > i have been generating a single shell model from the segmented mri. > the localisation proceedure is for a tactile stimulus so when i see a > contralateral S1 response i have some reason to belive that the algorithm is > making some sense. I get the impression from your reply that there is no > easy way to get around this lack of anatomical data and that i may just have > to go and get the mri's done for this couple of subjects! > > thanks for your thoughts on this, > > tom. > > > On 1/7/08, Jan Mathijs Schoffelen wrote: > > > > Dear Tom, > > > > As far as I know, the 'ctf-format' standardised mri does not exist. > > Creating an anatomically accurate forward model, while the actual anatomical > > information is lacking, is tricky business. What kind of forward model do > > you actully use? The multiple sphere model based on the ctf-shape files? If > > so, you could try to come up with an 'averaged' multipsphere model (thus > > bypassing the need for a standardised MRI), although this is not > > trial. However, I wonder whether this would yield better results than just > > taking an 'averaged' single sphere model. > > I wonder what you mean by your statement that localisation is much more > > accurate when using ctf-formatted MRIs. Do you know where your activation > > should end up then? > > > > The function beamformer_pcc is as far as I know still in its > > experimental stage and only for use within the FC Donders Centre. However, > > beamformer_dics shares most of the functionality, but obviously not all of > > it. > > > > Regards, > > > > Jan-Mathijs > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > > the FieldTrip toolbox, to share experiences and to discuss new ideas for > > MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and > EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > > > -------------------------------- > *Dr. Nathan Weisz* > > INSERM - Unité 821 > *Dynamique cérébrale et cognition* > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > *Tel*: ++33 - (0)4 - 7213 8915 > *Email: *weisz at lyon.inserm.fr > *Chat-AV*: nathanweisz at mac.com > *Homepage*: http://web.mac.com/nathanweisz > *Neurotree*: http://neurotree.org/neurotree/tree.php?pid=8692 > > > *Please avoid sending me Word or PowerPoint attachments.** > **See http://www.gnu.org/philosophy/no-word-attachments.html* > > > > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and > EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From linpe at NINDS.NIH.GOV Tue Jan 8 21:34:37 2008 From: linpe at NINDS.NIH.GOV (Peter Lin) Date: Tue, 8 Jan 2008 21:34:37 +0100 Subject: DICS pcc error Message-ID: I have a question. We are trying to process data to use DICS using fieldtrip at NIH using CTF system. In using the pcc script for preprocessing, we get this error: ??? Error using ==> ctf2grad at 68 unexpected number of coils in MEG channel Error in ==> read_header at 305 hdr.grad = ctf2grad(orig); Error in ==> read_fcdc_header at 37 [hdr] = read_header(varargin{:}); Error in ==> preprocessing at 291 hdr = read_fcdc_header(cfg.headerfile); In looking at the MEG channels, there are just the standard MEG ref channels, EEG channels (6) and MEG channels. Not sure where the program is going wrong. this is 20070204 version of fieldtrip. Thanks, peter Lin NINDS/NIH ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Wed Jan 9 08:23:45 2008 From: jan.schoffelen at FCDONDERS.RU.NL (Jan Mathijs Schoffelen) Date: Wed, 9 Jan 2008 08:23:45 +0100 Subject: DICS pcc error Message-ID: Dear Peter, First, I would advice you to update your fieldtrip-version to a more recent one. Apparently something is going wrong in read_fcdc_header, and not in the beamformer-algorithm. It seems as if the gradiometer-information in your headerfile (which is the .res4-file in the ctf-dataset) does not contain the information which fieldTrip expects. You might have to look into that. One likely cause is the absence of head-position information in your dataset. As a consequence the gradiometer-info in your .res4-file will only be specified in dewar-coordinates, which makes the dataset useless for source localisation anyway, because the MEG-data cannot be aligned to an anatomical coordinate system. Yours, Jan-Mathijs ----- Original Message ----- From: Peter Lin Date: Tuesday, January 8, 2008 9:34 pm Subject: [FIELDTRIP] DICS pcc error > I have a question. We are trying to process data to use DICS using > fieldtrip at NIH using CTF system. In using the pcc script for > preprocessing, we get this error: > > ??? Error using ==> ctf2grad at 68 > unexpected number of coils in MEG channel > > Error in ==> read_header at 305 > hdr.grad = ctf2grad(orig); > > Error in ==> read_fcdc_header at 37 > [hdr] = read_header(varargin{:}); > > Error in ==> preprocessing at 291 > hdr = read_fcdc_header(cfg.headerfile); > > In looking at the MEG channels, there are just the standard MEG ref > channels, EEG channels (6) and MEG channels. Not sure where the > program is > going wrong. this is 20070204 version of fieldtrip. > > Thanks, > > peter Lin > NINDS/NIH > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jedmeltzer at YAHOO.COM Thu Jan 10 00:34:34 2008 From: jedmeltzer at YAHOO.COM (Jed Meltzer) Date: Wed, 9 Jan 2008 15:34:34 -0800 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: Message-ID: The error may possibly have something to with the fact that our CTF system at NIH is missing two channels (due to broken squids) - maybe the software expects the full 275 to be there. I don't know though, as I have successfully imported 273-channel data from that system into fieldtrip, although I had other problems past the import stage that I have not gotten around to troubleshooting... -Jed --- FIELDTRIP automatic digest system wrote: --------------------------------- FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Service Desk : LISTSERV Archives FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Table of contents: DICS pcc error DICS pcc error Re: DICS pcc error (01/09) From: Jan Mathijs Schoffelen --------------------------------- Browse the FIELDTRIP online archives.> Date: Wed, 9 Jan 2008 08:23:45 +0100 > From: Jan Mathijs Schoffelen > > Subject: Re: DICS pcc error > > Dear Peter, > > First, I would advice you to update your > fieldtrip-version to a more recent one. > > Apparently something is going wrong in > read_fcdc_header, and not in the > beamformer-algorithm. It seems as if the > gradiometer-information in your headerfile (which is > the .res4-file in the ctf-dataset) does not contain > the information which fieldTrip expects. You might > have to look into that. One likely cause is the > absence of head-position information in your > dataset. As a consequence the gradiometer-info in > your .res4-file will only be specified in > dewar-coordinates, which makes the dataset useless > for source localisation anyway, because the MEG-data > cannot be aligned to an anatomical coordinate > system. > > Yours, > > Jan-Mathijs > > > ----- Original Message ----- > From: Peter Lin > Date: Tuesday, January 8, 2008 9:34 pm > Subject: [FIELDTRIP] DICS pcc error > > > I have a question. We are trying to process data > to use DICS using > > fieldtrip at NIH using CTF system. In using the > pcc script for > > preprocessing, we get this error: > > > > ??? Error using ==> ctf2grad at 68 > > unexpected number of coils in MEG channel > > > > Error in ==> read_header at 305 > > hdr.grad = ctf2grad(orig); > > > > Error in ==> read_fcdc_header at 37 > > [hdr] = read_header(varargin{:}); > > > > Error in ==> preprocessing at 291 > > hdr = read_fcdc_header(cfg.headerfile); > > > > In looking at the MEG channels, there are just the > standard MEG ref > > channels, EEG channels (6) and MEG channels. Not > sure where the > > program is > > going wrong. this is 20070204 version of > fieldtrip. > > > > Thanks, > > > > peter Lin > > NINDS/NIH > > > > ---------------------------------- > > The aim of this list is to facilitate the > discussion between users > > of the FieldTrip toolbox, to share experiences > and to discuss new > > ideas for MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html > and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > The aim of this list is to facilitate the discussion > between users of the FieldTrip toolbox, to share > experiences and to discuss new ideas for MEG and EEG > analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From linpe at NINDS.NIH.GOV Thu Jan 10 01:47:44 2008 From: linpe at NINDS.NIH.GOV (Lin, Peter (NIH/NINDS) [E]) Date: Wed, 9 Jan 2008 19:47:44 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Message-ID: HI Jed, thanks. Does Tom know? Peter -----Original Message----- From: Jed Meltzer [mailto:jedmeltzer at YAHOO.COM] Sent: Wed 1/9/2008 6:34 PM To: FIELDTRIP at NIC.SURFNET.NL Subject: Re: [FIELDTRIP] FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) The error may possibly have something to with the fact that our CTF system at NIH is missing two channels (due to broken squids) - maybe the software expects the full 275 to be there. I don't know though, as I have successfully imported 273-channel data from that system into fieldtrip, although I had other problems past the import stage that I have not gotten around to troubleshooting... -Jed --- FIELDTRIP automatic digest system wrote: --------------------------------- FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Service Desk : LISTSERV Archives FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Table of contents: DICS pcc error DICS pcc error Re: DICS pcc error (01/09) From: Jan Mathijs Schoffelen --------------------------------- Browse the FIELDTRIP online archives.> Date: Wed, 9 Jan 2008 08:23:45 +0100 > From: Jan Mathijs Schoffelen > > Subject: Re: DICS pcc error > > Dear Peter, > > First, I would advice you to update your > fieldtrip-version to a more recent one. > > Apparently something is going wrong in > read_fcdc_header, and not in the > beamformer-algorithm. It seems as if the > gradiometer-information in your headerfile (which is > the .res4-file in the ctf-dataset) does not contain > the information which fieldTrip expects. You might > have to look into that. One likely cause is the > absence of head-position information in your > dataset. As a consequence the gradiometer-info in > your .res4-file will only be specified in > dewar-coordinates, which makes the dataset useless > for source localisation anyway, because the MEG-data > cannot be aligned to an anatomical coordinate > system. > > Yours, > > Jan-Mathijs > > > ----- Original Message ----- > From: Peter Lin > Date: Tuesday, January 8, 2008 9:34 pm > Subject: [FIELDTRIP] DICS pcc error > > > I have a question. We are trying to process data > to use DICS using > > fieldtrip at NIH using CTF system. In using the > pcc script for > > preprocessing, we get this error: > > > > ??? Error using ==> ctf2grad at 68 > > unexpected number of coils in MEG channel > > > > Error in ==> read_header at 305 > > hdr.grad = ctf2grad(orig); > > > > Error in ==> read_fcdc_header at 37 > > [hdr] = read_header(varargin{:}); > > > > Error in ==> preprocessing at 291 > > hdr = read_fcdc_header(cfg.headerfile); > > > > In looking at the MEG channels, there are just the > standard MEG ref > > channels, EEG channels (6) and MEG channels. Not > sure where the > > program is > > going wrong. this is 20070204 version of > fieldtrip. > > > > Thanks, > > > > peter Lin > > NINDS/NIH > > > > ---------------------------------- > > The aim of this list is to facilitate the > discussion between users > > of the FieldTrip toolbox, to share experiences > and to discuss new > > ideas for MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html > and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > The aim of this list is to facilitate the discussion > between users of the FieldTrip toolbox, to share > experiences and to discuss new ideas for MEG and EEG > analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From venug001 at BAMA.UA.EDU Thu Jan 10 20:00:13 2008 From: venug001 at BAMA.UA.EDU (Gopa Venugopalan) Date: Thu, 10 Jan 2008 20:00:13 +0100 Subject: Importing 4-D datasets Message-ID: Hello, I wish everyone a productive New Year! Does anyone have any experience importing 4-D data? I have 146 channels of MEG and 25 channels of EEG that I am importing. That data has been preprocessed and grand averaged using proprretary 4-D software. All I now need is to import the ASCII files containing the averaged data. Thanking everyone in advance for any and all assistance. Warm Regards gopa ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 10 23:06:31 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 10 Jan 2008 17:06:31 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <265F73253EB9234F81FF4EC2DCB2AD1CA8B25B@nihcesmlbx2.nih.gov> Message-ID: On Wed, 2008-01-09 at 19:47 -0500, Lin, Peter (NIH/NINDS) [E] wrote: > HI Jed, > > thanks. Does Tom know? Tom is working on it. All I know at this point is that I tried it with Octave 3.0 (which just came out recently) and it works in Octave. Fieldtrip comes up with way too many markers in the stim channel for some reason under recent matlab, but it's OK in Octave. (Ha!) Dr. Tom -- My brother, when you have a virtue, and it is your own virtue, you have it in common with no one. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 09:41:26 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 09:41:26 +0100 Subject: Importing 4-D datasets In-Reply-To: Message-ID: hi gopa, the 4-D software should come with a command line tool called pdf2set. files exported with this command can be read in a standard fieldtrip manner. yet this assumes you want to read trials. you already averaged the data and probably want a timelockanalysis structure. perhaps it is the least pain to read the data (exported via pdf2set) with eeglab. then do dummy = eeglab2fieldtrip(EEG, 'timelockanalysis'). best wishes, nathan On 10.01.2008, at 20:00, Gopa Venugopalan wrote: > Hello, I wish everyone a productive New Year! Does anyone have any > experience importing 4-D data? I have 146 channels of MEG and 25 > channels > of EEG that I am importing. That data has been preprocessed and grand > averaged using proprretary 4-D software. All I now need is to import > the > ASCII files containing the averaged data. Thanking everyone in > advance for > any and all assistance. > Warm Regards > gopa > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ewingerd at SCIENCE.UVA.NL Fri Jan 11 10:22:04 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Fri, 11 Jan 2008 10:22:04 +0100 Subject: non-uniform trial length In-Reply-To: Message-ID: Hi Fieldtrippers, Recently, I've begun trying to use FieldTrip to analyse LFP traces from rats implanted with tetrode arrays. I would like to look at changes in power over a range of frequencies in relation to behavioural events. So, likely, I will use freqanalysis_tfr as a method for this. Since I'm interested in /changes/ in power from baseline, I would probably have to use freqbaseline with the 'relative change' method. Looking at the code for this function, it seems that it expects a baseline period somewhere before the event of interest, for each trial. In my research, the rats are freely moving around a cage, so there is no fixed interval between the end of the baseline period and the start of the peri-event interval of interest. As I see it, the baselinefreq function requires the baseline period to be a fixed part of any trial. Is there already a way to pass a full, pre-made freq.powspctrm (from the baseline periods) to the freqbaseline function? I noticed that the subfunction TFrelchange indexes into the event intervals. I guess I could rewrite that part as a private function to accept such a baseline freq.powspctrm, but I'm wondering if such a tool already exists. Alternatively, I guess I could create a 'trialfun' function that stitches the baseline and event part of the data together on a trial-by-trial basis, resulting in a uniform trial length. I would greatly appreciate any feedback on whether to go with A or B, or which solution would work the best in the FieldTrip framework, thanks, marijn -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 11:54:06 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 11:54:06 +0100 Subject: non-uniform trial length In-Reply-To: <4787353C.8020605@science.uva.nl> Message-ID: Dear Marjin, i don't know if it's useful for you but i adapted the freqbaseline function (attached). using cfg.otherbl you can pass on a freqanalysis another structure which will then be used to baseline-correct your freqanalysis structure of interest (see the help). if you decide to test it, I would be interested if it works for you. also I have a colleague who would also like to analyse rat data with Fieldtrip. I'd be interested in your opinion if its well-suited. btw, for time-frequency stuff i have good experiences with freqanalysis_wltconvol. all the best, nathan On 11.01.2008, at 10:22, Marijn van Wingerden wrote: > Marijn van Wingerden -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: freqbaseline.m Type: application/octet-stream Size: 13658 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: From venug001 at BAMA.UA.EDU Fri Jan 11 13:03:07 2008 From: venug001 at BAMA.UA.EDU (Gopakumar Venugopalan) Date: Fri, 11 Jan 2008 06:03:07 -0600 Subject: Importing 4-D datasets In-Reply-To: <159CABC0-0AB1-4572-A4E7-38160064A0AC@lyon.inserm.fr> Message-ID: Thanks Nathan. I did install pdf2set. The 4-D data is not discrimnating between EEG and MEG as the EEG electrodes were not digitized. So it is exported as one large 171 channel file. Yes, I guess timelock structure is where I want to start from then move to Topographic analysis, areal analysis and dipole fitting. Would you happen to know how many files pdf2set will export and with what extensions? I have ascii files of the data sets, I saw an .xyz file (cartesian coordinates) and an elp file (electrode position) file. Can I use the last two files in conjunction with the standard 10-20 electrode system position to read in the EEG data? Thanks, once again, in advance. Warm regards gopa Quoting Nathan Weisz : > hi gopa, > > the 4-D software should come with a command line tool called pdf2set. > > files exported with this command can be read in a standard fieldtrip > > manner. yet this assumes you want to read trials. > you already averaged the data and probably want a timelockanalysis > structure. perhaps it is the least pain to read the data (exported > via > pdf2set) with eeglab. then do > dummy = eeglab2fieldtrip(EEG, 'timelockanalysis'). > > best wishes, > nathan > > > > > > > On 10.01.2008, at 20:00, Gopa Venugopalan wrote: > > > Hello, I wish everyone a productive New Year! Does anyone have any > > experience importing 4-D data? I have 146 channels of MEG and 25 > > channels > > of EEG that I am importing. That data has been preprocessed and > grand > > averaged using proprretary 4-D software. All I now need is to > import > > the > > ASCII files containing the averaged data. Thanking everyone in > > advance for > > any and all assistance. > > Warm Regards > > gopa > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users > > > of the FieldTrip toolbox, to share experiences and to discuss new > > > ideas for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > > and http://www.ru.nl/fcdonders/fieldtrip. > > -------------------------------- > Dr. Nathan Weisz > > INSERM - Unité 821 > Dynamique cérébrale et cognition > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > Tel: ++33 - (0)4 - 7213 8915 > Email: weisz at lyon.inserm.fr > Chat-AV: nathanweisz at mac.com > Homepage: http://web.mac.com/nathanweisz > Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 > > Please avoid sending me Word or PowerPoint attachments. > See http://www.gnu.org/philosophy/no-word-attachments.html > > > > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From l.frei at PSY.GLA.AC.UK Fri Jan 11 13:47:44 2008 From: l.frei at PSY.GLA.AC.UK (Luisa Frei) Date: Fri, 11 Jan 2008 12:47:44 +0000 Subject: missing files, reshape Message-ID: Hi, I’m currently doing the fieldtrip tutorials and I have encountered 2 problems: 1. Some of the pre-calculated data files are missing. That’s especially complicating for the Parametric and non-parametric statistics tutorial, because the whole tutorial relies on that file: GA_ERF_orig.mat I also don't have the files for the cluster-based permutation test tutorial. Does anyone know, where I can find them? 2. In the cluster-based permutation test tutorial I get an error with the reshape function at the end of 2.1.2.1. Here’s the error text: >> [stat] = timelockstatistics(cfg, FICdata, FCdata); ??? Error using ==> reshape To RESHAPE the number of elements must not change. Error in ==> fieldtrip-20080107/private/ prepare_timefreq_data>forcedimord at 674 output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); Error in ==> timelockstatistics at 107 [stat] = statistics_wrapper(cfg, varargin{:}); I didn’t manage to fix this. Has anyone else encountered this and knows how to solve it? Any help would be greatly appreciated. Thanks, Luisa ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 14:28:40 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 14:28:40 +0100 Subject: Importing 4-D datasets In-Reply-To: <1200052987.47875afbab7fa@bamamail.ua.edu> Message-ID: hi gopa, > between EEG and MEG as the EEG electrodes were not digitized. So it is > exported as one large 171 channel file. using: hdr=read_fcdc_header('yourm4dfile.m4d') and then: hdr.grad.label you shoud see which sensors / electrodes correspond to which row in your data matrix. > Would you happen to know how many files pdf2set will export and with > what extensions? I have ascii files of the data sets, I saw an .xyz > file (cartesian coordinates) and an elp file (electrode position) > file. next to the data and the xyz you should have the header-file (m4d). if you want to do dipole fits you also need the headshape file (hs_file; read it via read_bti_hs). [just a side note: if it's a simple dipole fit you want then the 4d software should be good enough, so why go through the pain of exporting and importing data] > Can I use the last two files in conjunction with the standard 10-20 > electrode system position to read in the EEG data? I am not quite sure if i understand you correctly. but many fieldtrip functions (e.g. definetrial, timelockanalysis, freqanalysis) allow you to specifiy the channels you want, e.g.: cfg.channel={'MEG'}; you may try: cfg.channel={'EEG'}; and see if it only applies the relevant functions to the EEG channels (i haven't done this myself). if not you can specify all labels by hand or add something to the channelselction.m function that suits your needs. regarding electrode positions on the fieldtrip-FTP (standard_BEM) there are standard 10-20 and 10-05 elc files. you can read those files using read_fcdc_elec. hope this helps, nathan -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ewingerd at SCIENCE.UVA.NL Tue Jan 15 09:51:03 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Tue, 15 Jan 2008 09:51:03 +0100 Subject: non-uniform trial length In-Reply-To: Message-ID: Dear Nathan, I think that is exactly the fix I was looking for. I'll let you know what comes out of testing the .m file thanks a lot, all the best, marijn Nathan Weisz wrote: > Dear Marjin, > > i don't know if it's useful for you but i adapted the freqbaseline > function (attached). > using cfg.otherbl > you can pass on a freqanalysis another structure which will then be > used to baseline-correct your freqanalysis structure of interest (see > the help). > > if you decide to test it, I would be interested if it works for you. > also I have a colleague who would also like to analyse rat data with > Fieldtrip. I'd be interested in your opinion if its well-suited. > > btw, for time-frequency stuff i have good experiences > with freqanalysis_wltconvol. > > all the best, > nathan > > > > On 11.01.2008, at 10:22, Marijn van Wingerden wrote: > >> Marijn van Wingerden > > -------------------------------- > *Dr. Nathan Weisz* > > INSERM - Unité 821 > /Dynamique cérébrale et cognition/ > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > *Tel*: ++33 - (0)4 - 7213 8915 > *Email: *weisz at lyon.inserm.fr > *Chat-AV*: nathanweisz at mac.com > *Homepage*: http://web.mac.com/nathanweisz > *Neurotree*: http://neurotree.org/neurotree/tree.php?pid=8692 > > /Please avoid sending me Word or PowerPoint attachments.// > //See http://www.gnu.org/philosophy/no-word-attachments.html/ > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > > > ------------------------------------------------------------------------ > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Tue Jan 15 12:20:25 2008 From: jan.schoffelen at FCDONDERS.RU.NL (jan-mathijs schoffelen) Date: Tue, 15 Jan 2008 12:20:25 +0100 Subject: missing files, reshape In-Reply-To: <4FE97248-C603-452A-AA08-ED56D7E18EBB@PSY.GLA.AC.UK> Message-ID: Dear Luisa, I put the missing GA_ERF_orig mat-file on the ftp-server. You can access it at: ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail address). It's in the directory /pub/temporary. As far as I can see, the second problem could be related to the fact that you do not have the correct data. It might be solved when you use the data in the mat-file. Yours, Jan-Mathijs On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > Hi, > I’m currently doing the fieldtrip tutorials and I have encountered > 2 problems: > > 1. Some of the pre-calculated data files are missing. That’s > especially complicating for the Parametric and non-parametric > statistics tutorial, because the whole tutorial relies on that > file: GA_ERF_orig.mat > I also don't have the files for the cluster-based permutation test > tutorial. Does anyone know, where I can find them? > > 2. In the cluster-based permutation test tutorial I get an error > with the reshape function at the end of 2.1.2.1. Here’s the error > text: > > >> [stat] = timelockstatistics(cfg, FICdata, FCdata); > ??? Error using ==> reshape > To RESHAPE the number of elements must not change. > > Error in ==> fieldtrip-20080107/private/ > prepare_timefreq_data>forcedimord at 674 > output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); > > Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 > [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); > > Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 > [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); > > Error in ==> timelockstatistics at 107 > [stat] = statistics_wrapper(cfg, varargin{:}); > > > I didn’t manage to fix this. Has anyone else encountered this and > knows how to solve it? Any help would be greatly appreciated. > > > Thanks, > Luisa > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From ingrid.nieuwenhuis at FCDONDERS.RU.NL Tue Jan 15 12:41:59 2008 From: ingrid.nieuwenhuis at FCDONDERS.RU.NL (Ingrid Nieuwenhuis) Date: Tue, 15 Jan 2008 12:41:59 +0100 Subject: missing files, reshape In-Reply-To: Message-ID: Dear Louisa, The second problem might also be related to the dof field. If there's a dof field in the data, removing it might be a fix. Best Ingrid -----Original Message----- From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf Of jan-mathijs schoffelen Sent: Tuesday, January 15, 2008 12:20 PM To: FIELDTRIP at NIC.SURFNET.NL Subject: Re: [FIELDTRIP] missing files, reshape Dear Luisa, I put the missing GA_ERF_orig mat-file on the ftp-server. You can access it at: ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail address). It's in the directory /pub/temporary. As far as I can see, the second problem could be related to the fact that you do not have the correct data. It might be solved when you use the data in the mat-file. Yours, Jan-Mathijs On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > Hi, > I'm currently doing the fieldtrip tutorials and I have encountered > 2 problems: > > 1. Some of the pre-calculated data files are missing. That's > especially complicating for the Parametric and non-parametric > statistics tutorial, because the whole tutorial relies on that > file: GA_ERF_orig.mat > I also don't have the files for the cluster-based permutation test > tutorial. Does anyone know, where I can find them? > > 2. In the cluster-based permutation test tutorial I get an error > with the reshape function at the end of 2.1.2.1. Here's the error > text: > > >> [stat] = timelockstatistics(cfg, FICdata, FCdata); > ??? Error using ==> reshape > To RESHAPE the number of elements must not change. > > Error in ==> fieldtrip-20080107/private/ > prepare_timefreq_data>forcedimord at 674 > output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); > > Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 > [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); > > Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 > [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); > > Error in ==> timelockstatistics at 107 > [stat] = statistics_wrapper(cfg, varargin{:}); > > > I didn't manage to fix this. Has anyone else encountered this and > knows how to solve it? Any help would be greatly appreciated. > > > Thanks, > Luisa > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From r.oostenveld at FCDONDERS.RU.NL Wed Jan 16 09:11:52 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Wed, 16 Jan 2008 09:11:52 +0100 Subject: Importing 4-D datasets In-Reply-To: <02BA90E7-72D6-4BA1-9DF6-45C678A1F60D@lyon.inserm.fr> Message-ID: Hi Gopa Regarding combined EEG and MEG data, of course we also have this often with our CTF MEG system. The sensor locations stored in a preprocessed combined EEG+MEG data structure at the moment can not contain mixed EEG and MEG positions. Usually you will want to do source reconstruction with the MEG channels. In that case the data (after preprocessing, timelockanalysis or freqanalysis or componentanalysis) should have the field data.grad. If you want to do sourceanalysis/dipolefitting on the EEG channels, you should remove the data.grad field and instead insert a field into the data with the name data.elec. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes _magnetometers_or_gradiometers_described for fuether details. On 11 Jan 2008, at 14:28, Nathan Weisz wrote: > you may try: > cfg.channel={'EEG'}; > and see if it only applies the relevant functions to the EEG > channels (i haven't done this myself). if not you can specify all > labels by hand or add something to the channelselction.m function > that suits your needs. Usefull here might also be to specify cfg.channel='gui' and a graphical interface will pop up allowing you to select channels with the mouse. best regards, Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From venug001 at BAMA.UA.EDU Wed Jan 16 13:36:27 2008 From: venug001 at BAMA.UA.EDU (Gopakumar Venugopalan) Date: Wed, 16 Jan 2008 06:36:27 -0600 Subject: Importing 4-D datasets In-Reply-To: Message-ID: Greeting Robert, thanks for the freeware and assistance. Nathan has been very,very helpful from the outset. I am truly grateful to all out there best regards gopa Quoting Robert Oostenveld : > Hi Gopa > > Regarding combined EEG and MEG data, of course we also have this > often with our CTF MEG system. The sensor locations stored in a > preprocessed combined EEG+MEG data structure at the moment can not > contain mixed EEG and MEG positions. Usually you will want to do > source reconstruction with the MEG channels. In that case the data > (after preprocessing, timelockanalysis or freqanalysis or > componentanalysis) should have the field data.grad. If you want to do > > sourceanalysis/dipolefitting on the EEG channels, you should remove > > the data.grad field and instead insert a field into the data with the > > name data.elec. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? > > id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes > > _magnetometers_or_gradiometers_described for fuether details. > > On 11 Jan 2008, at 14:28, Nathan Weisz wrote: > > you may try: > > cfg.channel={'EEG'}; > > and see if it only applies the relevant functions to the EEG > > channels (i haven't done this myself). if not you can specify all > > > labels by hand or add something to the channelselction.m function > > > that suits your needs. > > Usefull here might also be to specify cfg.channel='gui' and a > graphical interface will pop up allowing you to select channels with > > the mouse. > > best regards, > Robert > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From r.oostenveld at FCDONDERS.RU.NL Wed Jan 16 17:32:14 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Wed, 16 Jan 2008 17:32:14 +0100 Subject: topoplot error In-Reply-To: Message-ID: Hi Thomas, On 6 Jan 2008, at 18:40, thomas gilbertson wrote: > Does anyone know where to find the standardised montreal " average" > mri in ctf format? Maybe a bit late, but I'll give a short reply anyway. If you read the canonical SPM MRI using mri=read_fcdc_mri(..), it will have mri.transform describing the homogenous coordinate transformation matrix that converts from voxel indices into SPM coordinates. In volumenormalise around line 153, there is a homogenous transformation matrix that converts an MRI from CTF head coordinates (linked to NAS/LPA/RPA) into SPM coordinates (linked to AC/PC). If you multiply the inverse of that 4x4 matrix with mri.transform that you obtained after reading the SPM canonical MRI, you should get the canonical MRI reasonably aligned with the CTF headcoordinate system. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? id=fieldtrip:documentation:frequently_asked_questions#how_are_the_differ ent_head_and_mri_coordinate_systems_defined for further details on coordinate systems, and the transformation matrix. best regards Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Wed Jan 16 21:13:20 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Wed, 16 Jan 2008 15:13:20 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <265F73253EB9234F81FF4EC2DCB2AD1CA8B25B@nihcesmlbx2.nih.gov> Message-ID: > > ----- Original Message ----- > > From: Peter Lin > > Date: Tuesday, January 8, 2008 9:34 pm > > Subject: [FIELDTRIP] DICS pcc error > > > > > > ??? Error using ==> ctf2grad at 68 > > > unexpected number of coils in MEG channel OK, the ultimate cause of the problem is line 143 in ctf_read_res4.m, which reads: temp=fread(fid,32,'char')'; it should be changed to: temp=fread(fid,32,'uchar')'; and that'll fix it. It is interesting that this works in Octave but not in real Matlab. What happens in Matlab with 'char' is that temp comes back as an array of _shorts_. I have no idea at all why this happens, it's completely wrong behavior. I'd appreciate it if somebody could check me on this. I'm using Matlab 7.2.0.294. With 'uchar' it works as it should. What happens with 'char', in more detail, is that it reads shorts, and the file pointer gets thrown off, and the gains, which are read in next, get read as zero. Also the coil numbers. Dr. Tom -- To them sounds it pleasant to have preached in their ears: "Nothing is worth while! You shall not will!" That, however, is a sermon for slavery. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From hubert.preissl at UNI-TUEBINGEN.DE Thu Jan 17 16:08:26 2008 From: hubert.preissl at UNI-TUEBINGEN.DE (Hubert Preissl) Date: Thu, 17 Jan 2008 16:08:26 +0100 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <1200514400.5186.74.camel@kurage.nimh.nih.gov> Message-ID: hi tom, the error is related to a change in char definition: http://www.mathworks.com/support/solutions/data/1-2782ZU.html?solution=1-2782ZU hubert -- Dr. Hubert Preissl MEG-Center phone: ++49-(0)7071-2987704 Otfried Mueller Str 47 fax: ++49-(0)7071-295706 72076 Tübingen, Germany email: hubert.preissl at uni-tuebingen.de ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From h.morgan at BANGOR.AC.UK Thu Jan 17 18:21:08 2008 From: h.morgan at BANGOR.AC.UK (Helen Morgan) Date: Thu, 17 Jan 2008 18:21:08 +0100 Subject: Post-doc position at Bangor University, UK Message-ID: Research Officer Wolfson Centre For Clinical And Cognitive Neuroscience School of Psychology Starting Salary: £21,478 - £23,883 (on R&A Grade 1A) p.a. Applications are invited for a three-year Postdoctoral Research Officer position in social cognitive neuroscience. This position is to work on a research grant awarded to Professor David Linden jointly with Professor Thomas Mussweiler of the University of Cologne, Germany. This grant is funded by the ESRC through their bilateral programme with the German Research Council (DFG). The aim of the project is to investigate the neural substrates of social comparison, one of the most pervasive cognitive processes humans engage in. The work will involve the adaptation of paradigms from social psychology research for functional imaging and the acquisition and analysis of behavioural and imaging data. The successful candidate will work closely with the research officer in the Cologne team, who is being recruited in parallel. Professor Mussweiler's group in Cologne is internationally leading in the fields of social judgement, stereotyping and decision making: http://social-cognition.uni-koeln.de/. This is a full-time position, starting on April 1st 2008. The ideal candidate would have a PhD in cognitive neuroscience or a related field, experience with stimulus programming and functional magnetic resonance imaging and an interest in the application of these skills to current topics in social psychology. The School of Psychology at Bangor is one of the UK's leading Psychology departments. It enjoys the highest possible research rating (RAE 5*A), and offers an exciting research environment within a dynamic and growing imaging group. Further details of the School, including our extensive research facilities (3T MRI, ERP, MRI-guided TMS), can be found under http://www.psychology.bangor.ac.uk/. The University is situated on the northern coastline of Wales in an area of outstanding natural beauty within a few miles of the Snowdonia Mountains National Park and several quiet beaches. Bangor has direct rail connections to London (3 hours) and Manchester (2 hours). Application forms and further particulars should be obtained by contacting Human Resources, Bangor University; tel: (01248) 382926/388132; e-mail: personnel at bangor.ac.uk; web: www.bangor.ac.uk For an informal discussion about this post, potential applicants are invited to contact Professor David Linden, tel: (01248) 382564, e-mail: d.linden at bangor.ac.uk Please quote reference number 08-7/93 when applying. Closing date for applications: Friday 15th February, 2008. Committed To Equal Opportunities ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 17 20:28:23 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 17 Jan 2008 14:28:23 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <478F6F6A.7020007@uni-tuebingen.de> Message-ID: Thank you! That was very helpful. I posted on the Octave list that I thought it was great that Octave 3.0 had gotten it right while Matlab had gotten it wrong. Of course now Matlab is saying that a char is 16 bits, breaking old code. Normal people introduce a wchar type to handle 16 bit characters. I hope the Octave people do something sensible, like provide "broken matlab mode" but leave 'char' as 8 bits by default. On Thu, 2008-01-17 at 16:08 +0100, Hubert Preissl wrote: > hi tom, > the error is related to a change in char definition: > > http://www.mathworks.com/support/solutions/data/1-2782ZU.html?solution=1-2782ZU > > > hubert > Dr. Tom -- He has subdued monsters, he has solved enigmas. But he should also redeem his monsters and enigmas; into heavenly children should he transform them. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 17 21:00:50 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 17 Jan 2008 15:00:50 -0500 Subject: char vs uchar Message-ID: The Octave guru suggests that it's better to use 'uint8' as that is unlikely to change. Dr. Tom -- Better, however, to be foolish with happiness than foolish with misfortune, better to dance awkwardly than walk lamely. So learn, I pray you, my wisdom, you higher men: even the worst thing has two good reverse sides, even the worst thing has good dancing-legs: so learn, I pray you, you higher men, to put yourselves on your proper legs! Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From l.frei at PSY.GLA.AC.UK Fri Jan 18 16:24:52 2008 From: l.frei at PSY.GLA.AC.UK (Luisa Frei) Date: Fri, 18 Jan 2008 15:24:52 +0000 Subject: missing files, reshape In-Reply-To: <003001c8576b$a3bfc3b0$642dae83@fcdonders.nl> Message-ID: Ingrid: removing the dof-field did he trick! Jan-Mathijs: thanks for the file! Many thanks to both of you for your help! Luisa On 15 Jan 2008, at 11:41, Ingrid Nieuwenhuis wrote: > Dear Louisa, > > The second problem might also be related to the dof field. If > there's a dof > field in the data, removing it might be a fix. > > Best Ingrid > > -----Original Message----- > From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] > On Behalf > Of jan-mathijs schoffelen > Sent: Tuesday, January 15, 2008 12:20 PM > To: FIELDTRIP at NIC.SURFNET.NL > Subject: Re: [FIELDTRIP] missing files, reshape > > Dear Luisa, > > I put the missing GA_ERF_orig mat-file on the ftp-server. You can > access it at: > > ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail > address). It's in the directory /pub/temporary. > > As far as I can see, the second problem could be related to the fact > that you do not have the correct data. It might be solved when you > use the data in the mat-file. > > Yours, > > Jan-Mathijs > > > > On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > >> Hi, >> I'm currently doing the fieldtrip tutorials and I have encountered >> 2 problems: >> >> 1. Some of the pre-calculated data files are missing. That's >> especially complicating for the Parametric and non-parametric >> statistics tutorial, because the whole tutorial relies on that >> file: GA_ERF_orig.mat >> I also don't have the files for the cluster-based permutation test >> tutorial. Does anyone know, where I can find them? >> >> 2. In the cluster-based permutation test tutorial I get an error >> with the reshape function at the end of 2.1.2.1. Here's the error >> text: >> >>>> [stat] = timelockstatistics(cfg, FICdata, FCdata); >> ??? Error using ==> reshape >> To RESHAPE the number of elements must not change. >> >> Error in ==> fieldtrip-20080107/private/ >> prepare_timefreq_data>forcedimord at 674 >> output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); >> >> Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 >> [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); >> >> Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 >> [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); >> >> Error in ==> timelockstatistics at 107 >> [stat] = statistics_wrapper(cfg, varargin{:}); >> >> >> I didn't manage to fix this. Has anyone else encountered this and >> knows how to solve it? Any help would be greatly appreciated. >> >> >> Thanks, >> Luisa >> ---------------------------------- >> The aim of this list is to facilitate the discussion between users >> of the FieldTrip toolbox, to share experiences and to discuss new >> ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ >> archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. >> > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the > FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 21 19:12:05 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 21 Jan 2008 18:12:05 +0000 Subject: topoplot error In-Reply-To: <4B8237E1-3717-431E-A772-D0D94DDDBC03@fcdonders.ru.nl> Message-ID: hi robert, a belated reply. appologies in advance if i have made some really simple mistake here but ive tryed what you suggested and am struggling to implement it successfully. heres what ive done so far:- % here is the variable to multiply mri.tranform with (from volumenormalise) initial = [ 0.0000 -1.0000 0.0000 0.0000 0.9987 0.0000 -0.0517 -32.0000 0.0517 0.0000 0.9987 -54.0000 0.0000 0.0000 0.0000 1.0000 ]; % mri is the imported mri mri.transform = 2 0 0 -92 0 2 0 -128 0 0 2 -74 0 0 0 1 % multiply with the inverse of initial and include in the interpolated source array interp.transform = mri.transform.*initial'; % run volume normalise cfg=[]; cfg.coordinates = 'ctf'; interpN = volumenormalise(cfg, interp); % and i get the following error; ??? Error using ==> colon Maximum variable size allowed by the program is exceeded. Error in ==> spm_smoothto8bit>smoothto8bit at 36 x = -r{i}.s:r{i}.s; ive truncated the error for brevity. thanks for your patience with this. best wishes, tom. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From linpe at NINDS.NIH.GOV Thu Jan 24 03:14:26 2008 From: linpe at NINDS.NIH.GOV (Peter Lin) Date: Thu, 24 Jan 2008 03:14:26 +0100 Subject: DICS error Message-ID: Recently, there was a problem with running fieldtrip analysis off of the NIMH MEG data for DICS. This worked after a change of ctf_read_res4.m file. I ran the DICS script and it was fine until this error came up: the input is freq data with 273 channels, 21 frequencybins and no timebins reading headmodel from file /home/linpe/raid/linpe/20071031/**_pright.ds/default.hdm using gradiometers specified in the data 3794 dipoles inside, 11074 dipoles outside brain computing cross-spectrum from fourier ??? Reference to non-existent field 'refchan'. Error in ==> prepare_freq_matrices at 247 refindx = match_str(freq.label, cfg.refchan); Error in ==> sourceanalysis at 714 [Cf, Cr, Pr, Ntrials, cfg] = prepare_freq_matrices(cfg, data); Error in ==> pcc1 at 61 [source] = sourceanalysis(cfg, freq); Any idea what led to this? Thanks in advance, Peter ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tillmann at MPIH-FRANKFURT.MPG.DE Fri Jan 25 09:53:31 2008 From: tillmann at MPIH-FRANKFURT.MPG.DE (Christine Tillmann) Date: Fri, 25 Jan 2008 09:53:31 +0100 Subject: Sourcestatistics & MNI-grids Message-ID: Hi All, I'm trying to compute virtual channel data using the lcmv-beamformer. As a first step, I created MNI-aligned dipole grids for all subjects to be able to combine the output of source analysis across subjects. But now it seems a bit tricky to feed those source-level data into sourcestatistics: Sourcestatistics seems to require computing the grand average over subjects first, but using sourcegrandaverage here returns the error message that grid locations are different across subjects. Do I need to interpolate and normalize each volume after all, before combining the source-level data? Maybe by using the transformation matrix that was computed for each subject when I normalised each individual anatomy to the MNI-template?? Or else, by copying the template grid positions to the individual subject data directly?? Can anyone help me out? I'd be grateful for any advice on this! Thanks & Best Regards, Christine ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tillmann at MPIH-FRANKFURT.MPG.DE Fri Jan 25 16:25:41 2008 From: tillmann at MPIH-FRANKFURT.MPG.DE (Christine Tillmann) Date: Fri, 25 Jan 2008 16:25:41 +0100 Subject: virtual channels Message-ID: Hi everyone, I got rid of the problem I described just before (combining source-level data in sourcestatistics with different grid positions across subjects) by copying the template grid structure to the output of sourceanalysis for each subject, before computing sourcegrandaverage and then doing sourcestatistics. Now, since I want to compute virtual channel data, I'm looking for a way to match the location coordinates of the activated voxels to the filter weights. Does anybody know how to match the indices of specific grid points given in 'inside' and 'filter' to the voxel coordinates given by sourceplot? Best regards, Christine ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From grion at SISSA.IT Mon Jan 28 13:34:14 2008 From: grion at SISSA.IT (Natalia Grion) Date: Mon, 28 Jan 2008 13:34:14 +0100 Subject: Importing Data from Neuralynx Message-ID: Hi All, I want to start with Fieldtrip analysis on animal data recorded with Neuralynx. I have LFP (*.ncs files), spikes (*.nts), and event files (*.nev). So, I have the three wrapper functions; which call these read_xxx functions that detect the file format and subsequently call the appropriate low-level function for each file format. In a first attempt to import data I ran: hdr = read_fcdc_header(file) with a *.ncs file but I found this error: ??? Undefined command/function 'read_neuralynx_ncs'. Error in ==> fieldtrip-20070820\private\read_header at 472 ncs = read_neuralynx_ncs(filename, 1, 0); Error in ==> read_fcdc_header at 37 [hdr] = read_header(varargin{:}); Do you have any explanation for this error? Another question: Are *.nst (spike files) recognized? Thank you all!! ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From ewingerd at SCIENCE.UVA.NL Mon Jan 28 14:17:55 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Mon, 28 Jan 2008 14:17:55 +0100 Subject: Importing Data from Neuralynx In-Reply-To: <479DCBC6.6090203@sissa.it> Message-ID: Hello Natalia, I've been working on getting Neuralynx data into Fieldtrip as well. There are some small and some large problems involved, especially concerning the registration of events. Feel free to e-mail me off this list with specific problems. It seems this particular error occured because you don't have the proper loading functions that the wrapper function calls in your path. They are located in the fileio subfolder, but I'll attach the relevant Nlx functions here in a zip-file. Just put them in your matlab path and the function should be able to run. good luck! marijn van wingerden Natalia Grion wrote: > Hi All, I want to start with Fieldtrip analysis on animal data > recorded with Neuralynx. I have LFP (*.ncs files), spikes (*.nts), and > event files (*.nev). So, I have the three wrapper functions; which > call these read_xxx functions that detect the file format and > subsequently call the appropriate low-level function for each file > format. > In a first attempt to import data I ran: > hdr = read_fcdc_header(file) with a *.ncs file > but I found this error: > ??? Undefined command/function 'read_neuralynx_ncs'. > Error in ==> fieldtrip-20070820\private\read_header at 472 > ncs = read_neuralynx_ncs(filename, 1, 0); > Error in ==> read_fcdc_header at 37 > [hdr] = read_header(varargin{:}); > > Do you have any explanation for this error? > Another question: Are *.nst (spike files) recognized? > > Thank you all!! > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- A non-text attachment was scrubbed... Name: read_neuralynx.zip Type: application/x-zip-compressed Size: 20484 bytes Desc: not available URL: From fredericroux at HOTMAIL.DE Fri Jan 4 11:04:03 2008 From: fredericroux at HOTMAIL.DE (Frederic Roux) Date: Fri, 4 Jan 2008 11:04:03 +0100 Subject: topoplot error Message-ID: Hi everybody, I've been trying to use the topoplotTFR function but constantly receive the error message 'Odd number of arguments?'. Has anyone ever encountered that problem? I used the following commands: cfg = []; cfg.layout = prepare_layout(data); cfg.baselinetype = 'relative'; cfg.xparam = 'time'; cfg.yparam = 'freq'; cfg.zparam = 'powspctrm'; cfg.xlim = 'maxmin'; cfg.ylim = 'maxmin'; cfg.zlim = 'maxmin'; topoplotTFR(cfg,data) topoplotTFR calls topoplotER which again calls topoplot. The problem seems to be related to the following lines of code: in topoplotER @ line 364: topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); and in topoplot @ line 957: if nargs> 2 if ~(round(nargs/2) == nargs/2) error('Odd number of input arguments??') end I don't get any further. Can anyone help me, please? Regards, Frederic _________________________________________________________________ Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! http://messenger.live.de/community/neuekontakte_adressimport.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From v.litvak at ION.UCL.AC.UK Fri Jan 4 11:24:35 2008 From: v.litvak at ION.UCL.AC.UK (Vladimir Litvak) Date: Fri, 4 Jan 2008 10:24:35 -0000 Subject: topoplot error In-Reply-To: Message-ID: This probably has to do with EEGLAB being above Fieldtrip in your Matlab path with EEGLAB's version of topoplot being called instead of Fieldtrip's. You need to put Fieldtrip on top and make sure it stays there because EEGLAB moves itself to the top each time it's called. Perhaps you should add a line modifying the path to your script. Best, Vladimir -----Original Message----- From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf Of Frederic Roux Sent: Friday, January 04, 2008 10:04 AM To: FIELDTRIP at NIC.SURFNET.NL Subject: [FIELDTRIP] topoplot error Hi everybody, I've been trying to use the topoplotTFR function but constantly receive the error message 'Odd number of arguments?'. Has anyone ever encountered that problem? I used the following commands: cfg = []; cfg.layout = prepare_layout(data); cfg.baselinetype = 'relative'; cfg.xparam = 'time'; cfg.yparam = 'freq'; cfg.zparam = 'powspctrm'; cfg.xlim = 'maxmin'; cfg.ylim = 'maxmin'; cfg.zlim = 'maxmin'; topoplotTFR(cfg,data) topoplotTFR calls topoplotER which again calls topoplot. The problem seems to be related to the following lines of code: in topoplotER @ line 364: topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); and in topoplot @ line 957: if nargs> 2 if ~(round(nargs/2) == nargs/2) error('Odd number of input arguments??') end I don't get any further. Can anyone help me, please? Regards, Frederic _________________________________________________________________ Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! http://messenger.live.de/community/neuekontakte_adressimport.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From fredericroux at HOTMAIL.DE Fri Jan 4 14:30:00 2008 From: fredericroux at HOTMAIL.DE (Frederic Roux) Date: Fri, 4 Jan 2008 14:30:00 +0100 Subject: topoplot error In-Reply-To: <001501c84ebc$071a4740$f0463ec1@sobell.ion.ucl.ac.uk> Message-ID: Hello Vladimir, thank you for the advice. That did it indeed. You saved my day :) ---------------------------------------- > Date: Fri, 4 Jan 2008 10:24:35 +0000 > From: v.litvak at ION.UCL.AC.UK > Subject: Re: [FIELDTRIP] topoplot error > To: FIELDTRIP at NIC.SURFNET.NL > > This probably has to do with EEGLAB being above Fieldtrip in your Matlab > path with EEGLAB's version of topoplot being called instead of Fieldtrip's. > You need to put Fieldtrip on top and make sure it stays there because EEGLAB > moves itself to the top each time it's called. Perhaps you should add a line > modifying the path to your script. > > Best, > > Vladimir > > -----Original Message----- > From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf > Of Frederic Roux > Sent: Friday, January 04, 2008 10:04 AM > To: FIELDTRIP at NIC.SURFNET.NL > Subject: [FIELDTRIP] topoplot error > > Hi everybody, > > I've been trying to use the topoplotTFR function but constantly receive the > error message 'Odd number of arguments?'. > Has anyone ever encountered that problem? > > I used the following commands: > > cfg = []; > cfg.layout = prepare_layout(data); > cfg.baselinetype = 'relative'; > cfg.xparam = 'time'; > cfg.yparam = 'freq'; > cfg.zparam = 'powspctrm'; > cfg.xlim = 'maxmin'; > cfg.ylim = 'maxmin'; > cfg.zlim = 'maxmin'; > > topoplotTFR(cfg,data) > > topoplotTFR calls topoplotER which again calls topoplot. > > The problem seems to be related to the following lines of code: > > in topoplotER @ line 364: > > topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); > > and in topoplot @ line 957: > > if nargs> 2 > if ~(round(nargs/2) == nargs/2) > error('Odd number of input arguments??') > end > > I don't get any further. Can anyone help me, please? > > Regards, > > Frederic > _________________________________________________________________ > Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! > http://messenger.live.de/community/neuekontakte_adressimport.html > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. _________________________________________________________________ Windows Live Fotogalerie: So einfach organisieren Sie Ihre Fotos! http://get.live.com/photogallery/overview ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Sun Jan 6 18:40:59 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Sun, 6 Jan 2008 17:40:59 +0000 Subject: topoplot error In-Reply-To: Message-ID: Hi everyone, Does anyone know where to find the standardised montreal " average" mri in ctf format? I see that there are some average mri's in the spm toolbox(spm2\canonical) but these are .mnc files in spm coordinates. I have two subjects without mris and have found that the localisation is much more accurate if i use another subjects ctf mri than if i use the mri's provided with spm (to produce the headmodel and for source interpolation). The MEG scanner used was a ctf machine so i thought that using a ctf mri average would give the best result. I have been careful to ensure that the coordinate system used both for the headmodel formation and for interpolation has been defined appropriately for the type of mri being used (eg cfg.coordinates = 'ctf'; or ='spm'). Any thoughts would be much appreciated. Also i cant seem to find the function beamformer_pcc.m which is called when using sourceanalysis.m, cfg. method ='pcc'; knowledge of its whereabouts would be greatly received! best wishes tom gilbertson ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at FCDONDERS.RU.NL Mon Jan 7 15:03:55 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Mon, 7 Jan 2008 15:03:55 +0100 Subject: automatic update wiki/ftp disabled for a few days Message-ID: Dear all The automatic updating of the documentation on the fieldtrip website and the automatic updating of the fieldtrip code on /home/common (internal to FC Donders) and on the external ftp server are TEMPORARY DISABLED. This is necessary due to migrating the cvs repository to a new server and to implementing new release management scripts. The automatic updating will be up again as soon as possible, but it might take a few days. best regards, Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Mon Jan 7 15:18:22 2008 From: jan.schoffelen at FCDONDERS.RU.NL (Jan Mathijs Schoffelen) Date: Mon, 7 Jan 2008 15:18:22 +0100 Subject: forward modelling without anatomical info Message-ID: Dear Tom, As far as I know, the 'ctf-format' standardised mri does not exist. Creating an anatomically accurate forward model, while the actual anatomical information is lacking, is tricky business. What kind of forward model do you actully use? The multiple sphere model based on the ctf-shape files? If so, you could try to come up with an 'averaged' multipsphere model (thus bypassing the need for a standardised MRI), although this is not trial. However, I wonder whether this would yield better results than just taking an 'averaged' single sphere model. I wonder what you mean by your statement that localisation is much more accurate when using ctf-formatted MRIs. Do you know where your activation should end up then? The function beamformer_pcc is as far as I know still in its experimental stage and only for use within the FC Donders Centre. However, beamformer_dics shares most of the functionality, but obviously not all of it. Regards, Jan-Mathijs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 7 16:59:09 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 7 Jan 2008 15:59:09 +0000 Subject: forward modelling without anatomical info In-Reply-To: <3c02ed1911ff6.478242be@ru.nl> Message-ID: dear jan mathijs, i have been generating a single shell model from the segmented mri. the localisation proceedure is for a tactile stimulus so when i see a contralateral S1 response i have some reason to belive that the algorithm is making some sense. I get the impression from your reply that there is no easy way to get around this lack of anatomical data and that i may just have to go and get the mri's done for this couple of subjects! thanks for your thoughts on this, tom. On 1/7/08, Jan Mathijs Schoffelen wrote: > > Dear Tom, > > As far as I know, the 'ctf-format' standardised mri does not exist. > Creating an anatomically accurate forward model, while the actual anatomical > information is lacking, is tricky business. What kind of forward model do > you actully use? The multiple sphere model based on the ctf-shape files? If > so, you could try to come up with an 'averaged' multipsphere model (thus > bypassing the need for a standardised MRI), although this is not > trial. However, I wonder whether this would yield better results than just > taking an 'averaged' single sphere model. > I wonder what you mean by your statement that localisation is much more > accurate when using ctf-formatted MRIs. Do you know where your activation > should end up then? > > The function beamformer_pcc is as far as I know still in its experimental > stage and only for use within the FC Donders Centre. However, > beamformer_dics shares most of the functionality, but obviously not all of > it. > > Regards, > > Jan-Mathijs > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From weisz at LYON.INSERM.FR Mon Jan 7 17:11:57 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Mon, 7 Jan 2008 17:11:57 +0100 Subject: forward modelling without anatomical info In-Reply-To: Message-ID: hi, on the fieldtrip FTP there is a standard BEM (standard_vol.matr). i think you could use the skull part as pseudo-headshape: vol.bnd(1).pnt this also fits with the standard MRI. there may be some transformation you have to apply to these "headshape" points so that the head fits nicely into the sensor array. then you can use prepare_localspheres and specify a lower number in cfg.spheremesh (perhaps something like 50; default is 4000). this may yield a crude but satisfactory solution if you are unable to get MRIs or individual headshapes. i'd be interested in the outcome. all the best, nathan On 07.01.2008, at 16:59, thomas gilbertson wrote: > dear jan mathijs, > > i have been generating a single shell model from the segmented mri. > the localisation proceedure is for a tactile stimulus so when i see > a contralateral S1 response i have some reason to belive that the > algorithm is making some sense. I get the impression from your reply > that there is no easy way to get around this lack of anatomical data > and that i may just have to go and get the mri's done for this > couple of subjects! > > thanks for your thoughts on this, > > tom. > > > On 1/7/08, Jan Mathijs Schoffelen > wrote: > Dear Tom, > > As far as I know, the 'ctf-format' standardised mri does not exist. > Creating an anatomically accurate forward model, while the actual > anatomical information is lacking, is tricky business. What kind of > forward model do you actully use? The multiple sphere model based on > the ctf-shape files? If so, you could try to come up with an > 'averaged' multipsphere model (thus bypassing the need for a > standardised MRI), although this is not trial. However, I wonder > whether this would yield better results than just taking an > 'averaged' single sphere model. > I wonder what you mean by your statement that localisation is much > more accurate when using ctf-formatted MRIs. Do you know where your > activation should end up then? > > The function beamformer_pcc is as far as I know still in its > experimental stage and only for use within the FC Donders Centre. > However, beamformer_dics shares most of the functionality, but > obviously not all of it. > > Regards, > > Jan-Mathijs > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 7 17:17:11 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 7 Jan 2008 16:17:11 +0000 Subject: forward modelling without anatomical info In-Reply-To: <99082A9C-EC3C-4EDE-884F-5D3D5AD3FADB@lyon.inserm.fr> Message-ID: thanks nathan, ill give it a try! tom On 1/7/08, Nathan Weisz wrote: > > hi, > > > on the fieldtrip FTP there is a standard BEM (standard_vol.matr). i think > you could use the skull part as pseudo-headshape: > vol.bnd(1).pnt > > > this also fits with the standard MRI. > > > there may be some transformation you have to apply to these "headshape" > points so that the head fits nicely into the sensor array. > then you can use prepare_localspheres and specify a lower number in > cfg.spheremesh (perhaps something like 50; default is 4000). > > > this may yield a crude but satisfactory solution if you are unable to get > MRIs or individual headshapes. > > > i'd be interested in the outcome. > > > all the best, > nathan > > > > > > > > On 07.01.2008, at 16:59, thomas gilbertson wrote: > > dear jan mathijs, > > i have been generating a single shell model from the segmented mri. > the localisation proceedure is for a tactile stimulus so when i see a > contralateral S1 response i have some reason to belive that the algorithm is > making some sense. I get the impression from your reply that there is no > easy way to get around this lack of anatomical data and that i may just have > to go and get the mri's done for this couple of subjects! > > thanks for your thoughts on this, > > tom. > > > On 1/7/08, Jan Mathijs Schoffelen wrote: > > > > Dear Tom, > > > > As far as I know, the 'ctf-format' standardised mri does not exist. > > Creating an anatomically accurate forward model, while the actual anatomical > > information is lacking, is tricky business. What kind of forward model do > > you actully use? The multiple sphere model based on the ctf-shape files? If > > so, you could try to come up with an 'averaged' multipsphere model (thus > > bypassing the need for a standardised MRI), although this is not > > trial. However, I wonder whether this would yield better results than just > > taking an 'averaged' single sphere model. > > I wonder what you mean by your statement that localisation is much more > > accurate when using ctf-formatted MRIs. Do you know where your activation > > should end up then? > > > > The function beamformer_pcc is as far as I know still in its > > experimental stage and only for use within the FC Donders Centre. However, > > beamformer_dics shares most of the functionality, but obviously not all of > > it. > > > > Regards, > > > > Jan-Mathijs > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > > the FieldTrip toolbox, to share experiences and to discuss new ideas for > > MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and > EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > > > -------------------------------- > *Dr. Nathan Weisz* > > INSERM - Unité 821 > *Dynamique cérébrale et cognition* > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > *Tel*: ++33 - (0)4 - 7213 8915 > *Email: *weisz at lyon.inserm.fr > *Chat-AV*: nathanweisz at mac.com > *Homepage*: http://web.mac.com/nathanweisz > *Neurotree*: http://neurotree.org/neurotree/tree.php?pid=8692 > > > *Please avoid sending me Word or PowerPoint attachments.** > **See http://www.gnu.org/philosophy/no-word-attachments.html* > > > > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and > EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From linpe at NINDS.NIH.GOV Tue Jan 8 21:34:37 2008 From: linpe at NINDS.NIH.GOV (Peter Lin) Date: Tue, 8 Jan 2008 21:34:37 +0100 Subject: DICS pcc error Message-ID: I have a question. We are trying to process data to use DICS using fieldtrip at NIH using CTF system. In using the pcc script for preprocessing, we get this error: ??? Error using ==> ctf2grad at 68 unexpected number of coils in MEG channel Error in ==> read_header at 305 hdr.grad = ctf2grad(orig); Error in ==> read_fcdc_header at 37 [hdr] = read_header(varargin{:}); Error in ==> preprocessing at 291 hdr = read_fcdc_header(cfg.headerfile); In looking at the MEG channels, there are just the standard MEG ref channels, EEG channels (6) and MEG channels. Not sure where the program is going wrong. this is 20070204 version of fieldtrip. Thanks, peter Lin NINDS/NIH ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Wed Jan 9 08:23:45 2008 From: jan.schoffelen at FCDONDERS.RU.NL (Jan Mathijs Schoffelen) Date: Wed, 9 Jan 2008 08:23:45 +0100 Subject: DICS pcc error Message-ID: Dear Peter, First, I would advice you to update your fieldtrip-version to a more recent one. Apparently something is going wrong in read_fcdc_header, and not in the beamformer-algorithm. It seems as if the gradiometer-information in your headerfile (which is the .res4-file in the ctf-dataset) does not contain the information which fieldTrip expects. You might have to look into that. One likely cause is the absence of head-position information in your dataset. As a consequence the gradiometer-info in your .res4-file will only be specified in dewar-coordinates, which makes the dataset useless for source localisation anyway, because the MEG-data cannot be aligned to an anatomical coordinate system. Yours, Jan-Mathijs ----- Original Message ----- From: Peter Lin Date: Tuesday, January 8, 2008 9:34 pm Subject: [FIELDTRIP] DICS pcc error > I have a question. We are trying to process data to use DICS using > fieldtrip at NIH using CTF system. In using the pcc script for > preprocessing, we get this error: > > ??? Error using ==> ctf2grad at 68 > unexpected number of coils in MEG channel > > Error in ==> read_header at 305 > hdr.grad = ctf2grad(orig); > > Error in ==> read_fcdc_header at 37 > [hdr] = read_header(varargin{:}); > > Error in ==> preprocessing at 291 > hdr = read_fcdc_header(cfg.headerfile); > > In looking at the MEG channels, there are just the standard MEG ref > channels, EEG channels (6) and MEG channels. Not sure where the > program is > going wrong. this is 20070204 version of fieldtrip. > > Thanks, > > peter Lin > NINDS/NIH > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jedmeltzer at YAHOO.COM Thu Jan 10 00:34:34 2008 From: jedmeltzer at YAHOO.COM (Jed Meltzer) Date: Wed, 9 Jan 2008 15:34:34 -0800 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: Message-ID: The error may possibly have something to with the fact that our CTF system at NIH is missing two channels (due to broken squids) - maybe the software expects the full 275 to be there. I don't know though, as I have successfully imported 273-channel data from that system into fieldtrip, although I had other problems past the import stage that I have not gotten around to troubleshooting... -Jed --- FIELDTRIP automatic digest system wrote: --------------------------------- FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Service Desk : LISTSERV Archives FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Table of contents: DICS pcc error DICS pcc error Re: DICS pcc error (01/09) From: Jan Mathijs Schoffelen --------------------------------- Browse the FIELDTRIP online archives.> Date: Wed, 9 Jan 2008 08:23:45 +0100 > From: Jan Mathijs Schoffelen > > Subject: Re: DICS pcc error > > Dear Peter, > > First, I would advice you to update your > fieldtrip-version to a more recent one. > > Apparently something is going wrong in > read_fcdc_header, and not in the > beamformer-algorithm. It seems as if the > gradiometer-information in your headerfile (which is > the .res4-file in the ctf-dataset) does not contain > the information which fieldTrip expects. You might > have to look into that. One likely cause is the > absence of head-position information in your > dataset. As a consequence the gradiometer-info in > your .res4-file will only be specified in > dewar-coordinates, which makes the dataset useless > for source localisation anyway, because the MEG-data > cannot be aligned to an anatomical coordinate > system. > > Yours, > > Jan-Mathijs > > > ----- Original Message ----- > From: Peter Lin > Date: Tuesday, January 8, 2008 9:34 pm > Subject: [FIELDTRIP] DICS pcc error > > > I have a question. We are trying to process data > to use DICS using > > fieldtrip at NIH using CTF system. In using the > pcc script for > > preprocessing, we get this error: > > > > ??? Error using ==> ctf2grad at 68 > > unexpected number of coils in MEG channel > > > > Error in ==> read_header at 305 > > hdr.grad = ctf2grad(orig); > > > > Error in ==> read_fcdc_header at 37 > > [hdr] = read_header(varargin{:}); > > > > Error in ==> preprocessing at 291 > > hdr = read_fcdc_header(cfg.headerfile); > > > > In looking at the MEG channels, there are just the > standard MEG ref > > channels, EEG channels (6) and MEG channels. Not > sure where the > > program is > > going wrong. this is 20070204 version of > fieldtrip. > > > > Thanks, > > > > peter Lin > > NINDS/NIH > > > > ---------------------------------- > > The aim of this list is to facilitate the > discussion between users > > of the FieldTrip toolbox, to share experiences > and to discuss new > > ideas for MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html > and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > The aim of this list is to facilitate the discussion > between users of the FieldTrip toolbox, to share > experiences and to discuss new ideas for MEG and EEG > analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From linpe at NINDS.NIH.GOV Thu Jan 10 01:47:44 2008 From: linpe at NINDS.NIH.GOV (Lin, Peter (NIH/NINDS) [E]) Date: Wed, 9 Jan 2008 19:47:44 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Message-ID: HI Jed, thanks. Does Tom know? Peter -----Original Message----- From: Jed Meltzer [mailto:jedmeltzer at YAHOO.COM] Sent: Wed 1/9/2008 6:34 PM To: FIELDTRIP at NIC.SURFNET.NL Subject: Re: [FIELDTRIP] FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) The error may possibly have something to with the fact that our CTF system at NIH is missing two channels (due to broken squids) - maybe the software expects the full 275 to be there. I don't know though, as I have successfully imported 273-channel data from that system into fieldtrip, although I had other problems past the import stage that I have not gotten around to troubleshooting... -Jed --- FIELDTRIP automatic digest system wrote: --------------------------------- FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Service Desk : LISTSERV Archives FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Table of contents: DICS pcc error DICS pcc error Re: DICS pcc error (01/09) From: Jan Mathijs Schoffelen --------------------------------- Browse the FIELDTRIP online archives.> Date: Wed, 9 Jan 2008 08:23:45 +0100 > From: Jan Mathijs Schoffelen > > Subject: Re: DICS pcc error > > Dear Peter, > > First, I would advice you to update your > fieldtrip-version to a more recent one. > > Apparently something is going wrong in > read_fcdc_header, and not in the > beamformer-algorithm. It seems as if the > gradiometer-information in your headerfile (which is > the .res4-file in the ctf-dataset) does not contain > the information which fieldTrip expects. You might > have to look into that. One likely cause is the > absence of head-position information in your > dataset. As a consequence the gradiometer-info in > your .res4-file will only be specified in > dewar-coordinates, which makes the dataset useless > for source localisation anyway, because the MEG-data > cannot be aligned to an anatomical coordinate > system. > > Yours, > > Jan-Mathijs > > > ----- Original Message ----- > From: Peter Lin > Date: Tuesday, January 8, 2008 9:34 pm > Subject: [FIELDTRIP] DICS pcc error > > > I have a question. We are trying to process data > to use DICS using > > fieldtrip at NIH using CTF system. In using the > pcc script for > > preprocessing, we get this error: > > > > ??? Error using ==> ctf2grad at 68 > > unexpected number of coils in MEG channel > > > > Error in ==> read_header at 305 > > hdr.grad = ctf2grad(orig); > > > > Error in ==> read_fcdc_header at 37 > > [hdr] = read_header(varargin{:}); > > > > Error in ==> preprocessing at 291 > > hdr = read_fcdc_header(cfg.headerfile); > > > > In looking at the MEG channels, there are just the > standard MEG ref > > channels, EEG channels (6) and MEG channels. Not > sure where the > > program is > > going wrong. this is 20070204 version of > fieldtrip. > > > > Thanks, > > > > peter Lin > > NINDS/NIH > > > > ---------------------------------- > > The aim of this list is to facilitate the > discussion between users > > of the FieldTrip toolbox, to share experiences > and to discuss new > > ideas for MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html > and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > The aim of this list is to facilitate the discussion > between users of the FieldTrip toolbox, to share > experiences and to discuss new ideas for MEG and EEG > analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From venug001 at BAMA.UA.EDU Thu Jan 10 20:00:13 2008 From: venug001 at BAMA.UA.EDU (Gopa Venugopalan) Date: Thu, 10 Jan 2008 20:00:13 +0100 Subject: Importing 4-D datasets Message-ID: Hello, I wish everyone a productive New Year! Does anyone have any experience importing 4-D data? I have 146 channels of MEG and 25 channels of EEG that I am importing. That data has been preprocessed and grand averaged using proprretary 4-D software. All I now need is to import the ASCII files containing the averaged data. Thanking everyone in advance for any and all assistance. Warm Regards gopa ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 10 23:06:31 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 10 Jan 2008 17:06:31 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <265F73253EB9234F81FF4EC2DCB2AD1CA8B25B@nihcesmlbx2.nih.gov> Message-ID: On Wed, 2008-01-09 at 19:47 -0500, Lin, Peter (NIH/NINDS) [E] wrote: > HI Jed, > > thanks. Does Tom know? Tom is working on it. All I know at this point is that I tried it with Octave 3.0 (which just came out recently) and it works in Octave. Fieldtrip comes up with way too many markers in the stim channel for some reason under recent matlab, but it's OK in Octave. (Ha!) Dr. Tom -- My brother, when you have a virtue, and it is your own virtue, you have it in common with no one. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 09:41:26 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 09:41:26 +0100 Subject: Importing 4-D datasets In-Reply-To: Message-ID: hi gopa, the 4-D software should come with a command line tool called pdf2set. files exported with this command can be read in a standard fieldtrip manner. yet this assumes you want to read trials. you already averaged the data and probably want a timelockanalysis structure. perhaps it is the least pain to read the data (exported via pdf2set) with eeglab. then do dummy = eeglab2fieldtrip(EEG, 'timelockanalysis'). best wishes, nathan On 10.01.2008, at 20:00, Gopa Venugopalan wrote: > Hello, I wish everyone a productive New Year! Does anyone have any > experience importing 4-D data? I have 146 channels of MEG and 25 > channels > of EEG that I am importing. That data has been preprocessed and grand > averaged using proprretary 4-D software. All I now need is to import > the > ASCII files containing the averaged data. Thanking everyone in > advance for > any and all assistance. > Warm Regards > gopa > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ewingerd at SCIENCE.UVA.NL Fri Jan 11 10:22:04 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Fri, 11 Jan 2008 10:22:04 +0100 Subject: non-uniform trial length In-Reply-To: Message-ID: Hi Fieldtrippers, Recently, I've begun trying to use FieldTrip to analyse LFP traces from rats implanted with tetrode arrays. I would like to look at changes in power over a range of frequencies in relation to behavioural events. So, likely, I will use freqanalysis_tfr as a method for this. Since I'm interested in /changes/ in power from baseline, I would probably have to use freqbaseline with the 'relative change' method. Looking at the code for this function, it seems that it expects a baseline period somewhere before the event of interest, for each trial. In my research, the rats are freely moving around a cage, so there is no fixed interval between the end of the baseline period and the start of the peri-event interval of interest. As I see it, the baselinefreq function requires the baseline period to be a fixed part of any trial. Is there already a way to pass a full, pre-made freq.powspctrm (from the baseline periods) to the freqbaseline function? I noticed that the subfunction TFrelchange indexes into the event intervals. I guess I could rewrite that part as a private function to accept such a baseline freq.powspctrm, but I'm wondering if such a tool already exists. Alternatively, I guess I could create a 'trialfun' function that stitches the baseline and event part of the data together on a trial-by-trial basis, resulting in a uniform trial length. I would greatly appreciate any feedback on whether to go with A or B, or which solution would work the best in the FieldTrip framework, thanks, marijn -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 11:54:06 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 11:54:06 +0100 Subject: non-uniform trial length In-Reply-To: <4787353C.8020605@science.uva.nl> Message-ID: Dear Marjin, i don't know if it's useful for you but i adapted the freqbaseline function (attached). using cfg.otherbl you can pass on a freqanalysis another structure which will then be used to baseline-correct your freqanalysis structure of interest (see the help). if you decide to test it, I would be interested if it works for you. also I have a colleague who would also like to analyse rat data with Fieldtrip. I'd be interested in your opinion if its well-suited. btw, for time-frequency stuff i have good experiences with freqanalysis_wltconvol. all the best, nathan On 11.01.2008, at 10:22, Marijn van Wingerden wrote: > Marijn van Wingerden -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: freqbaseline.m Type: application/octet-stream Size: 13658 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: From venug001 at BAMA.UA.EDU Fri Jan 11 13:03:07 2008 From: venug001 at BAMA.UA.EDU (Gopakumar Venugopalan) Date: Fri, 11 Jan 2008 06:03:07 -0600 Subject: Importing 4-D datasets In-Reply-To: <159CABC0-0AB1-4572-A4E7-38160064A0AC@lyon.inserm.fr> Message-ID: Thanks Nathan. I did install pdf2set. The 4-D data is not discrimnating between EEG and MEG as the EEG electrodes were not digitized. So it is exported as one large 171 channel file. Yes, I guess timelock structure is where I want to start from then move to Topographic analysis, areal analysis and dipole fitting. Would you happen to know how many files pdf2set will export and with what extensions? I have ascii files of the data sets, I saw an .xyz file (cartesian coordinates) and an elp file (electrode position) file. Can I use the last two files in conjunction with the standard 10-20 electrode system position to read in the EEG data? Thanks, once again, in advance. Warm regards gopa Quoting Nathan Weisz : > hi gopa, > > the 4-D software should come with a command line tool called pdf2set. > > files exported with this command can be read in a standard fieldtrip > > manner. yet this assumes you want to read trials. > you already averaged the data and probably want a timelockanalysis > structure. perhaps it is the least pain to read the data (exported > via > pdf2set) with eeglab. then do > dummy = eeglab2fieldtrip(EEG, 'timelockanalysis'). > > best wishes, > nathan > > > > > > > On 10.01.2008, at 20:00, Gopa Venugopalan wrote: > > > Hello, I wish everyone a productive New Year! Does anyone have any > > experience importing 4-D data? I have 146 channels of MEG and 25 > > channels > > of EEG that I am importing. That data has been preprocessed and > grand > > averaged using proprretary 4-D software. All I now need is to > import > > the > > ASCII files containing the averaged data. Thanking everyone in > > advance for > > any and all assistance. > > Warm Regards > > gopa > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users > > > of the FieldTrip toolbox, to share experiences and to discuss new > > > ideas for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > > and http://www.ru.nl/fcdonders/fieldtrip. > > -------------------------------- > Dr. Nathan Weisz > > INSERM - Unité 821 > Dynamique cérébrale et cognition > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > Tel: ++33 - (0)4 - 7213 8915 > Email: weisz at lyon.inserm.fr > Chat-AV: nathanweisz at mac.com > Homepage: http://web.mac.com/nathanweisz > Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 > > Please avoid sending me Word or PowerPoint attachments. > See http://www.gnu.org/philosophy/no-word-attachments.html > > > > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From l.frei at PSY.GLA.AC.UK Fri Jan 11 13:47:44 2008 From: l.frei at PSY.GLA.AC.UK (Luisa Frei) Date: Fri, 11 Jan 2008 12:47:44 +0000 Subject: missing files, reshape Message-ID: Hi, I’m currently doing the fieldtrip tutorials and I have encountered 2 problems: 1. Some of the pre-calculated data files are missing. That’s especially complicating for the Parametric and non-parametric statistics tutorial, because the whole tutorial relies on that file: GA_ERF_orig.mat I also don't have the files for the cluster-based permutation test tutorial. Does anyone know, where I can find them? 2. In the cluster-based permutation test tutorial I get an error with the reshape function at the end of 2.1.2.1. Here’s the error text: >> [stat] = timelockstatistics(cfg, FICdata, FCdata); ??? Error using ==> reshape To RESHAPE the number of elements must not change. Error in ==> fieldtrip-20080107/private/ prepare_timefreq_data>forcedimord at 674 output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); Error in ==> timelockstatistics at 107 [stat] = statistics_wrapper(cfg, varargin{:}); I didn’t manage to fix this. Has anyone else encountered this and knows how to solve it? Any help would be greatly appreciated. Thanks, Luisa ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 14:28:40 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 14:28:40 +0100 Subject: Importing 4-D datasets In-Reply-To: <1200052987.47875afbab7fa@bamamail.ua.edu> Message-ID: hi gopa, > between EEG and MEG as the EEG electrodes were not digitized. So it is > exported as one large 171 channel file. using: hdr=read_fcdc_header('yourm4dfile.m4d') and then: hdr.grad.label you shoud see which sensors / electrodes correspond to which row in your data matrix. > Would you happen to know how many files pdf2set will export and with > what extensions? I have ascii files of the data sets, I saw an .xyz > file (cartesian coordinates) and an elp file (electrode position) > file. next to the data and the xyz you should have the header-file (m4d). if you want to do dipole fits you also need the headshape file (hs_file; read it via read_bti_hs). [just a side note: if it's a simple dipole fit you want then the 4d software should be good enough, so why go through the pain of exporting and importing data] > Can I use the last two files in conjunction with the standard 10-20 > electrode system position to read in the EEG data? I am not quite sure if i understand you correctly. but many fieldtrip functions (e.g. definetrial, timelockanalysis, freqanalysis) allow you to specifiy the channels you want, e.g.: cfg.channel={'MEG'}; you may try: cfg.channel={'EEG'}; and see if it only applies the relevant functions to the EEG channels (i haven't done this myself). if not you can specify all labels by hand or add something to the channelselction.m function that suits your needs. regarding electrode positions on the fieldtrip-FTP (standard_BEM) there are standard 10-20 and 10-05 elc files. you can read those files using read_fcdc_elec. hope this helps, nathan -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ewingerd at SCIENCE.UVA.NL Tue Jan 15 09:51:03 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Tue, 15 Jan 2008 09:51:03 +0100 Subject: non-uniform trial length In-Reply-To: Message-ID: Dear Nathan, I think that is exactly the fix I was looking for. I'll let you know what comes out of testing the .m file thanks a lot, all the best, marijn Nathan Weisz wrote: > Dear Marjin, > > i don't know if it's useful for you but i adapted the freqbaseline > function (attached). > using cfg.otherbl > you can pass on a freqanalysis another structure which will then be > used to baseline-correct your freqanalysis structure of interest (see > the help). > > if you decide to test it, I would be interested if it works for you. > also I have a colleague who would also like to analyse rat data with > Fieldtrip. I'd be interested in your opinion if its well-suited. > > btw, for time-frequency stuff i have good experiences > with freqanalysis_wltconvol. > > all the best, > nathan > > > > On 11.01.2008, at 10:22, Marijn van Wingerden wrote: > >> Marijn van Wingerden > > -------------------------------- > *Dr. Nathan Weisz* > > INSERM - Unité 821 > /Dynamique cérébrale et cognition/ > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > *Tel*: ++33 - (0)4 - 7213 8915 > *Email: *weisz at lyon.inserm.fr > *Chat-AV*: nathanweisz at mac.com > *Homepage*: http://web.mac.com/nathanweisz > *Neurotree*: http://neurotree.org/neurotree/tree.php?pid=8692 > > /Please avoid sending me Word or PowerPoint attachments.// > //See http://www.gnu.org/philosophy/no-word-attachments.html/ > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > > > ------------------------------------------------------------------------ > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Tue Jan 15 12:20:25 2008 From: jan.schoffelen at FCDONDERS.RU.NL (jan-mathijs schoffelen) Date: Tue, 15 Jan 2008 12:20:25 +0100 Subject: missing files, reshape In-Reply-To: <4FE97248-C603-452A-AA08-ED56D7E18EBB@PSY.GLA.AC.UK> Message-ID: Dear Luisa, I put the missing GA_ERF_orig mat-file on the ftp-server. You can access it at: ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail address). It's in the directory /pub/temporary. As far as I can see, the second problem could be related to the fact that you do not have the correct data. It might be solved when you use the data in the mat-file. Yours, Jan-Mathijs On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > Hi, > I’m currently doing the fieldtrip tutorials and I have encountered > 2 problems: > > 1. Some of the pre-calculated data files are missing. That’s > especially complicating for the Parametric and non-parametric > statistics tutorial, because the whole tutorial relies on that > file: GA_ERF_orig.mat > I also don't have the files for the cluster-based permutation test > tutorial. Does anyone know, where I can find them? > > 2. In the cluster-based permutation test tutorial I get an error > with the reshape function at the end of 2.1.2.1. Here’s the error > text: > > >> [stat] = timelockstatistics(cfg, FICdata, FCdata); > ??? Error using ==> reshape > To RESHAPE the number of elements must not change. > > Error in ==> fieldtrip-20080107/private/ > prepare_timefreq_data>forcedimord at 674 > output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); > > Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 > [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); > > Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 > [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); > > Error in ==> timelockstatistics at 107 > [stat] = statistics_wrapper(cfg, varargin{:}); > > > I didn’t manage to fix this. Has anyone else encountered this and > knows how to solve it? Any help would be greatly appreciated. > > > Thanks, > Luisa > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From ingrid.nieuwenhuis at FCDONDERS.RU.NL Tue Jan 15 12:41:59 2008 From: ingrid.nieuwenhuis at FCDONDERS.RU.NL (Ingrid Nieuwenhuis) Date: Tue, 15 Jan 2008 12:41:59 +0100 Subject: missing files, reshape In-Reply-To: Message-ID: Dear Louisa, The second problem might also be related to the dof field. If there's a dof field in the data, removing it might be a fix. Best Ingrid -----Original Message----- From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf Of jan-mathijs schoffelen Sent: Tuesday, January 15, 2008 12:20 PM To: FIELDTRIP at NIC.SURFNET.NL Subject: Re: [FIELDTRIP] missing files, reshape Dear Luisa, I put the missing GA_ERF_orig mat-file on the ftp-server. You can access it at: ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail address). It's in the directory /pub/temporary. As far as I can see, the second problem could be related to the fact that you do not have the correct data. It might be solved when you use the data in the mat-file. Yours, Jan-Mathijs On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > Hi, > I'm currently doing the fieldtrip tutorials and I have encountered > 2 problems: > > 1. Some of the pre-calculated data files are missing. That's > especially complicating for the Parametric and non-parametric > statistics tutorial, because the whole tutorial relies on that > file: GA_ERF_orig.mat > I also don't have the files for the cluster-based permutation test > tutorial. Does anyone know, where I can find them? > > 2. In the cluster-based permutation test tutorial I get an error > with the reshape function at the end of 2.1.2.1. Here's the error > text: > > >> [stat] = timelockstatistics(cfg, FICdata, FCdata); > ??? Error using ==> reshape > To RESHAPE the number of elements must not change. > > Error in ==> fieldtrip-20080107/private/ > prepare_timefreq_data>forcedimord at 674 > output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); > > Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 > [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); > > Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 > [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); > > Error in ==> timelockstatistics at 107 > [stat] = statistics_wrapper(cfg, varargin{:}); > > > I didn't manage to fix this. Has anyone else encountered this and > knows how to solve it? Any help would be greatly appreciated. > > > Thanks, > Luisa > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From r.oostenveld at FCDONDERS.RU.NL Wed Jan 16 09:11:52 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Wed, 16 Jan 2008 09:11:52 +0100 Subject: Importing 4-D datasets In-Reply-To: <02BA90E7-72D6-4BA1-9DF6-45C678A1F60D@lyon.inserm.fr> Message-ID: Hi Gopa Regarding combined EEG and MEG data, of course we also have this often with our CTF MEG system. The sensor locations stored in a preprocessed combined EEG+MEG data structure at the moment can not contain mixed EEG and MEG positions. Usually you will want to do source reconstruction with the MEG channels. In that case the data (after preprocessing, timelockanalysis or freqanalysis or componentanalysis) should have the field data.grad. If you want to do sourceanalysis/dipolefitting on the EEG channels, you should remove the data.grad field and instead insert a field into the data with the name data.elec. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes _magnetometers_or_gradiometers_described for fuether details. On 11 Jan 2008, at 14:28, Nathan Weisz wrote: > you may try: > cfg.channel={'EEG'}; > and see if it only applies the relevant functions to the EEG > channels (i haven't done this myself). if not you can specify all > labels by hand or add something to the channelselction.m function > that suits your needs. Usefull here might also be to specify cfg.channel='gui' and a graphical interface will pop up allowing you to select channels with the mouse. best regards, Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From venug001 at BAMA.UA.EDU Wed Jan 16 13:36:27 2008 From: venug001 at BAMA.UA.EDU (Gopakumar Venugopalan) Date: Wed, 16 Jan 2008 06:36:27 -0600 Subject: Importing 4-D datasets In-Reply-To: Message-ID: Greeting Robert, thanks for the freeware and assistance. Nathan has been very,very helpful from the outset. I am truly grateful to all out there best regards gopa Quoting Robert Oostenveld : > Hi Gopa > > Regarding combined EEG and MEG data, of course we also have this > often with our CTF MEG system. The sensor locations stored in a > preprocessed combined EEG+MEG data structure at the moment can not > contain mixed EEG and MEG positions. Usually you will want to do > source reconstruction with the MEG channels. In that case the data > (after preprocessing, timelockanalysis or freqanalysis or > componentanalysis) should have the field data.grad. If you want to do > > sourceanalysis/dipolefitting on the EEG channels, you should remove > > the data.grad field and instead insert a field into the data with the > > name data.elec. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? > > id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes > > _magnetometers_or_gradiometers_described for fuether details. > > On 11 Jan 2008, at 14:28, Nathan Weisz wrote: > > you may try: > > cfg.channel={'EEG'}; > > and see if it only applies the relevant functions to the EEG > > channels (i haven't done this myself). if not you can specify all > > > labels by hand or add something to the channelselction.m function > > > that suits your needs. > > Usefull here might also be to specify cfg.channel='gui' and a > graphical interface will pop up allowing you to select channels with > > the mouse. > > best regards, > Robert > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From r.oostenveld at FCDONDERS.RU.NL Wed Jan 16 17:32:14 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Wed, 16 Jan 2008 17:32:14 +0100 Subject: topoplot error In-Reply-To: Message-ID: Hi Thomas, On 6 Jan 2008, at 18:40, thomas gilbertson wrote: > Does anyone know where to find the standardised montreal " average" > mri in ctf format? Maybe a bit late, but I'll give a short reply anyway. If you read the canonical SPM MRI using mri=read_fcdc_mri(..), it will have mri.transform describing the homogenous coordinate transformation matrix that converts from voxel indices into SPM coordinates. In volumenormalise around line 153, there is a homogenous transformation matrix that converts an MRI from CTF head coordinates (linked to NAS/LPA/RPA) into SPM coordinates (linked to AC/PC). If you multiply the inverse of that 4x4 matrix with mri.transform that you obtained after reading the SPM canonical MRI, you should get the canonical MRI reasonably aligned with the CTF headcoordinate system. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? id=fieldtrip:documentation:frequently_asked_questions#how_are_the_differ ent_head_and_mri_coordinate_systems_defined for further details on coordinate systems, and the transformation matrix. best regards Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Wed Jan 16 21:13:20 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Wed, 16 Jan 2008 15:13:20 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <265F73253EB9234F81FF4EC2DCB2AD1CA8B25B@nihcesmlbx2.nih.gov> Message-ID: > > ----- Original Message ----- > > From: Peter Lin > > Date: Tuesday, January 8, 2008 9:34 pm > > Subject: [FIELDTRIP] DICS pcc error > > > > > > ??? Error using ==> ctf2grad at 68 > > > unexpected number of coils in MEG channel OK, the ultimate cause of the problem is line 143 in ctf_read_res4.m, which reads: temp=fread(fid,32,'char')'; it should be changed to: temp=fread(fid,32,'uchar')'; and that'll fix it. It is interesting that this works in Octave but not in real Matlab. What happens in Matlab with 'char' is that temp comes back as an array of _shorts_. I have no idea at all why this happens, it's completely wrong behavior. I'd appreciate it if somebody could check me on this. I'm using Matlab 7.2.0.294. With 'uchar' it works as it should. What happens with 'char', in more detail, is that it reads shorts, and the file pointer gets thrown off, and the gains, which are read in next, get read as zero. Also the coil numbers. Dr. Tom -- To them sounds it pleasant to have preached in their ears: "Nothing is worth while! You shall not will!" That, however, is a sermon for slavery. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From hubert.preissl at UNI-TUEBINGEN.DE Thu Jan 17 16:08:26 2008 From: hubert.preissl at UNI-TUEBINGEN.DE (Hubert Preissl) Date: Thu, 17 Jan 2008 16:08:26 +0100 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <1200514400.5186.74.camel@kurage.nimh.nih.gov> Message-ID: hi tom, the error is related to a change in char definition: http://www.mathworks.com/support/solutions/data/1-2782ZU.html?solution=1-2782ZU hubert -- Dr. Hubert Preissl MEG-Center phone: ++49-(0)7071-2987704 Otfried Mueller Str 47 fax: ++49-(0)7071-295706 72076 Tübingen, Germany email: hubert.preissl at uni-tuebingen.de ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From h.morgan at BANGOR.AC.UK Thu Jan 17 18:21:08 2008 From: h.morgan at BANGOR.AC.UK (Helen Morgan) Date: Thu, 17 Jan 2008 18:21:08 +0100 Subject: Post-doc position at Bangor University, UK Message-ID: Research Officer Wolfson Centre For Clinical And Cognitive Neuroscience School of Psychology Starting Salary: £21,478 - £23,883 (on R&A Grade 1A) p.a. Applications are invited for a three-year Postdoctoral Research Officer position in social cognitive neuroscience. This position is to work on a research grant awarded to Professor David Linden jointly with Professor Thomas Mussweiler of the University of Cologne, Germany. This grant is funded by the ESRC through their bilateral programme with the German Research Council (DFG). The aim of the project is to investigate the neural substrates of social comparison, one of the most pervasive cognitive processes humans engage in. The work will involve the adaptation of paradigms from social psychology research for functional imaging and the acquisition and analysis of behavioural and imaging data. The successful candidate will work closely with the research officer in the Cologne team, who is being recruited in parallel. Professor Mussweiler's group in Cologne is internationally leading in the fields of social judgement, stereotyping and decision making: http://social-cognition.uni-koeln.de/. This is a full-time position, starting on April 1st 2008. The ideal candidate would have a PhD in cognitive neuroscience or a related field, experience with stimulus programming and functional magnetic resonance imaging and an interest in the application of these skills to current topics in social psychology. The School of Psychology at Bangor is one of the UK's leading Psychology departments. It enjoys the highest possible research rating (RAE 5*A), and offers an exciting research environment within a dynamic and growing imaging group. Further details of the School, including our extensive research facilities (3T MRI, ERP, MRI-guided TMS), can be found under http://www.psychology.bangor.ac.uk/. The University is situated on the northern coastline of Wales in an area of outstanding natural beauty within a few miles of the Snowdonia Mountains National Park and several quiet beaches. Bangor has direct rail connections to London (3 hours) and Manchester (2 hours). Application forms and further particulars should be obtained by contacting Human Resources, Bangor University; tel: (01248) 382926/388132; e-mail: personnel at bangor.ac.uk; web: www.bangor.ac.uk For an informal discussion about this post, potential applicants are invited to contact Professor David Linden, tel: (01248) 382564, e-mail: d.linden at bangor.ac.uk Please quote reference number 08-7/93 when applying. Closing date for applications: Friday 15th February, 2008. Committed To Equal Opportunities ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 17 20:28:23 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 17 Jan 2008 14:28:23 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <478F6F6A.7020007@uni-tuebingen.de> Message-ID: Thank you! That was very helpful. I posted on the Octave list that I thought it was great that Octave 3.0 had gotten it right while Matlab had gotten it wrong. Of course now Matlab is saying that a char is 16 bits, breaking old code. Normal people introduce a wchar type to handle 16 bit characters. I hope the Octave people do something sensible, like provide "broken matlab mode" but leave 'char' as 8 bits by default. On Thu, 2008-01-17 at 16:08 +0100, Hubert Preissl wrote: > hi tom, > the error is related to a change in char definition: > > http://www.mathworks.com/support/solutions/data/1-2782ZU.html?solution=1-2782ZU > > > hubert > Dr. Tom -- He has subdued monsters, he has solved enigmas. But he should also redeem his monsters and enigmas; into heavenly children should he transform them. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 17 21:00:50 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 17 Jan 2008 15:00:50 -0500 Subject: char vs uchar Message-ID: The Octave guru suggests that it's better to use 'uint8' as that is unlikely to change. Dr. Tom -- Better, however, to be foolish with happiness than foolish with misfortune, better to dance awkwardly than walk lamely. So learn, I pray you, my wisdom, you higher men: even the worst thing has two good reverse sides, even the worst thing has good dancing-legs: so learn, I pray you, you higher men, to put yourselves on your proper legs! Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From l.frei at PSY.GLA.AC.UK Fri Jan 18 16:24:52 2008 From: l.frei at PSY.GLA.AC.UK (Luisa Frei) Date: Fri, 18 Jan 2008 15:24:52 +0000 Subject: missing files, reshape In-Reply-To: <003001c8576b$a3bfc3b0$642dae83@fcdonders.nl> Message-ID: Ingrid: removing the dof-field did he trick! Jan-Mathijs: thanks for the file! Many thanks to both of you for your help! Luisa On 15 Jan 2008, at 11:41, Ingrid Nieuwenhuis wrote: > Dear Louisa, > > The second problem might also be related to the dof field. If > there's a dof > field in the data, removing it might be a fix. > > Best Ingrid > > -----Original Message----- > From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] > On Behalf > Of jan-mathijs schoffelen > Sent: Tuesday, January 15, 2008 12:20 PM > To: FIELDTRIP at NIC.SURFNET.NL > Subject: Re: [FIELDTRIP] missing files, reshape > > Dear Luisa, > > I put the missing GA_ERF_orig mat-file on the ftp-server. You can > access it at: > > ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail > address). It's in the directory /pub/temporary. > > As far as I can see, the second problem could be related to the fact > that you do not have the correct data. It might be solved when you > use the data in the mat-file. > > Yours, > > Jan-Mathijs > > > > On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > >> Hi, >> I'm currently doing the fieldtrip tutorials and I have encountered >> 2 problems: >> >> 1. Some of the pre-calculated data files are missing. That's >> especially complicating for the Parametric and non-parametric >> statistics tutorial, because the whole tutorial relies on that >> file: GA_ERF_orig.mat >> I also don't have the files for the cluster-based permutation test >> tutorial. Does anyone know, where I can find them? >> >> 2. In the cluster-based permutation test tutorial I get an error >> with the reshape function at the end of 2.1.2.1. Here's the error >> text: >> >>>> [stat] = timelockstatistics(cfg, FICdata, FCdata); >> ??? Error using ==> reshape >> To RESHAPE the number of elements must not change. >> >> Error in ==> fieldtrip-20080107/private/ >> prepare_timefreq_data>forcedimord at 674 >> output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); >> >> Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 >> [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); >> >> Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 >> [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); >> >> Error in ==> timelockstatistics at 107 >> [stat] = statistics_wrapper(cfg, varargin{:}); >> >> >> I didn't manage to fix this. Has anyone else encountered this and >> knows how to solve it? Any help would be greatly appreciated. >> >> >> Thanks, >> Luisa >> ---------------------------------- >> The aim of this list is to facilitate the discussion between users >> of the FieldTrip toolbox, to share experiences and to discuss new >> ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ >> archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. >> > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the > FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 21 19:12:05 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 21 Jan 2008 18:12:05 +0000 Subject: topoplot error In-Reply-To: <4B8237E1-3717-431E-A772-D0D94DDDBC03@fcdonders.ru.nl> Message-ID: hi robert, a belated reply. appologies in advance if i have made some really simple mistake here but ive tryed what you suggested and am struggling to implement it successfully. heres what ive done so far:- % here is the variable to multiply mri.tranform with (from volumenormalise) initial = [ 0.0000 -1.0000 0.0000 0.0000 0.9987 0.0000 -0.0517 -32.0000 0.0517 0.0000 0.9987 -54.0000 0.0000 0.0000 0.0000 1.0000 ]; % mri is the imported mri mri.transform = 2 0 0 -92 0 2 0 -128 0 0 2 -74 0 0 0 1 % multiply with the inverse of initial and include in the interpolated source array interp.transform = mri.transform.*initial'; % run volume normalise cfg=[]; cfg.coordinates = 'ctf'; interpN = volumenormalise(cfg, interp); % and i get the following error; ??? Error using ==> colon Maximum variable size allowed by the program is exceeded. Error in ==> spm_smoothto8bit>smoothto8bit at 36 x = -r{i}.s:r{i}.s; ive truncated the error for brevity. thanks for your patience with this. best wishes, tom. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From linpe at NINDS.NIH.GOV Thu Jan 24 03:14:26 2008 From: linpe at NINDS.NIH.GOV (Peter Lin) Date: Thu, 24 Jan 2008 03:14:26 +0100 Subject: DICS error Message-ID: Recently, there was a problem with running fieldtrip analysis off of the NIMH MEG data for DICS. This worked after a change of ctf_read_res4.m file. I ran the DICS script and it was fine until this error came up: the input is freq data with 273 channels, 21 frequencybins and no timebins reading headmodel from file /home/linpe/raid/linpe/20071031/**_pright.ds/default.hdm using gradiometers specified in the data 3794 dipoles inside, 11074 dipoles outside brain computing cross-spectrum from fourier ??? Reference to non-existent field 'refchan'. Error in ==> prepare_freq_matrices at 247 refindx = match_str(freq.label, cfg.refchan); Error in ==> sourceanalysis at 714 [Cf, Cr, Pr, Ntrials, cfg] = prepare_freq_matrices(cfg, data); Error in ==> pcc1 at 61 [source] = sourceanalysis(cfg, freq); Any idea what led to this? Thanks in advance, Peter ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tillmann at MPIH-FRANKFURT.MPG.DE Fri Jan 25 09:53:31 2008 From: tillmann at MPIH-FRANKFURT.MPG.DE (Christine Tillmann) Date: Fri, 25 Jan 2008 09:53:31 +0100 Subject: Sourcestatistics & MNI-grids Message-ID: Hi All, I'm trying to compute virtual channel data using the lcmv-beamformer. As a first step, I created MNI-aligned dipole grids for all subjects to be able to combine the output of source analysis across subjects. But now it seems a bit tricky to feed those source-level data into sourcestatistics: Sourcestatistics seems to require computing the grand average over subjects first, but using sourcegrandaverage here returns the error message that grid locations are different across subjects. Do I need to interpolate and normalize each volume after all, before combining the source-level data? Maybe by using the transformation matrix that was computed for each subject when I normalised each individual anatomy to the MNI-template?? Or else, by copying the template grid positions to the individual subject data directly?? Can anyone help me out? I'd be grateful for any advice on this! Thanks & Best Regards, Christine ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tillmann at MPIH-FRANKFURT.MPG.DE Fri Jan 25 16:25:41 2008 From: tillmann at MPIH-FRANKFURT.MPG.DE (Christine Tillmann) Date: Fri, 25 Jan 2008 16:25:41 +0100 Subject: virtual channels Message-ID: Hi everyone, I got rid of the problem I described just before (combining source-level data in sourcestatistics with different grid positions across subjects) by copying the template grid structure to the output of sourceanalysis for each subject, before computing sourcegrandaverage and then doing sourcestatistics. Now, since I want to compute virtual channel data, I'm looking for a way to match the location coordinates of the activated voxels to the filter weights. Does anybody know how to match the indices of specific grid points given in 'inside' and 'filter' to the voxel coordinates given by sourceplot? Best regards, Christine ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From grion at SISSA.IT Mon Jan 28 13:34:14 2008 From: grion at SISSA.IT (Natalia Grion) Date: Mon, 28 Jan 2008 13:34:14 +0100 Subject: Importing Data from Neuralynx Message-ID: Hi All, I want to start with Fieldtrip analysis on animal data recorded with Neuralynx. I have LFP (*.ncs files), spikes (*.nts), and event files (*.nev). So, I have the three wrapper functions; which call these read_xxx functions that detect the file format and subsequently call the appropriate low-level function for each file format. In a first attempt to import data I ran: hdr = read_fcdc_header(file) with a *.ncs file but I found this error: ??? Undefined command/function 'read_neuralynx_ncs'. Error in ==> fieldtrip-20070820\private\read_header at 472 ncs = read_neuralynx_ncs(filename, 1, 0); Error in ==> read_fcdc_header at 37 [hdr] = read_header(varargin{:}); Do you have any explanation for this error? Another question: Are *.nst (spike files) recognized? Thank you all!! ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From ewingerd at SCIENCE.UVA.NL Mon Jan 28 14:17:55 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Mon, 28 Jan 2008 14:17:55 +0100 Subject: Importing Data from Neuralynx In-Reply-To: <479DCBC6.6090203@sissa.it> Message-ID: Hello Natalia, I've been working on getting Neuralynx data into Fieldtrip as well. There are some small and some large problems involved, especially concerning the registration of events. Feel free to e-mail me off this list with specific problems. It seems this particular error occured because you don't have the proper loading functions that the wrapper function calls in your path. They are located in the fileio subfolder, but I'll attach the relevant Nlx functions here in a zip-file. Just put them in your matlab path and the function should be able to run. good luck! marijn van wingerden Natalia Grion wrote: > Hi All, I want to start with Fieldtrip analysis on animal data > recorded with Neuralynx. I have LFP (*.ncs files), spikes (*.nts), and > event files (*.nev). So, I have the three wrapper functions; which > call these read_xxx functions that detect the file format and > subsequently call the appropriate low-level function for each file > format. > In a first attempt to import data I ran: > hdr = read_fcdc_header(file) with a *.ncs file > but I found this error: > ??? Undefined command/function 'read_neuralynx_ncs'. > Error in ==> fieldtrip-20070820\private\read_header at 472 > ncs = read_neuralynx_ncs(filename, 1, 0); > Error in ==> read_fcdc_header at 37 > [hdr] = read_header(varargin{:}); > > Do you have any explanation for this error? > Another question: Are *.nst (spike files) recognized? > > Thank you all!! > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- A non-text attachment was scrubbed... Name: read_neuralynx.zip Type: application/x-zip-compressed Size: 20484 bytes Desc: not available URL: From fredericroux at HOTMAIL.DE Fri Jan 4 11:04:03 2008 From: fredericroux at HOTMAIL.DE (Frederic Roux) Date: Fri, 4 Jan 2008 11:04:03 +0100 Subject: topoplot error Message-ID: Hi everybody, I've been trying to use the topoplotTFR function but constantly receive the error message 'Odd number of arguments?'. Has anyone ever encountered that problem? I used the following commands: cfg = []; cfg.layout = prepare_layout(data); cfg.baselinetype = 'relative'; cfg.xparam = 'time'; cfg.yparam = 'freq'; cfg.zparam = 'powspctrm'; cfg.xlim = 'maxmin'; cfg.ylim = 'maxmin'; cfg.zlim = 'maxmin'; topoplotTFR(cfg,data) topoplotTFR calls topoplotER which again calls topoplot. The problem seems to be related to the following lines of code: in topoplotER @ line 364: topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); and in topoplot @ line 957: if nargs> 2 if ~(round(nargs/2) == nargs/2) error('Odd number of input arguments??') end I don't get any further. Can anyone help me, please? Regards, Frederic _________________________________________________________________ Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! http://messenger.live.de/community/neuekontakte_adressimport.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From v.litvak at ION.UCL.AC.UK Fri Jan 4 11:24:35 2008 From: v.litvak at ION.UCL.AC.UK (Vladimir Litvak) Date: Fri, 4 Jan 2008 10:24:35 -0000 Subject: topoplot error In-Reply-To: Message-ID: This probably has to do with EEGLAB being above Fieldtrip in your Matlab path with EEGLAB's version of topoplot being called instead of Fieldtrip's. You need to put Fieldtrip on top and make sure it stays there because EEGLAB moves itself to the top each time it's called. Perhaps you should add a line modifying the path to your script. Best, Vladimir -----Original Message----- From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf Of Frederic Roux Sent: Friday, January 04, 2008 10:04 AM To: FIELDTRIP at NIC.SURFNET.NL Subject: [FIELDTRIP] topoplot error Hi everybody, I've been trying to use the topoplotTFR function but constantly receive the error message 'Odd number of arguments?'. Has anyone ever encountered that problem? I used the following commands: cfg = []; cfg.layout = prepare_layout(data); cfg.baselinetype = 'relative'; cfg.xparam = 'time'; cfg.yparam = 'freq'; cfg.zparam = 'powspctrm'; cfg.xlim = 'maxmin'; cfg.ylim = 'maxmin'; cfg.zlim = 'maxmin'; topoplotTFR(cfg,data) topoplotTFR calls topoplotER which again calls topoplot. The problem seems to be related to the following lines of code: in topoplotER @ line 364: topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); and in topoplot @ line 957: if nargs> 2 if ~(round(nargs/2) == nargs/2) error('Odd number of input arguments??') end I don't get any further. Can anyone help me, please? Regards, Frederic _________________________________________________________________ Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! http://messenger.live.de/community/neuekontakte_adressimport.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From fredericroux at HOTMAIL.DE Fri Jan 4 14:30:00 2008 From: fredericroux at HOTMAIL.DE (Frederic Roux) Date: Fri, 4 Jan 2008 14:30:00 +0100 Subject: topoplot error In-Reply-To: <001501c84ebc$071a4740$f0463ec1@sobell.ion.ucl.ac.uk> Message-ID: Hello Vladimir, thank you for the advice. That did it indeed. You saved my day :) ---------------------------------------- > Date: Fri, 4 Jan 2008 10:24:35 +0000 > From: v.litvak at ION.UCL.AC.UK > Subject: Re: [FIELDTRIP] topoplot error > To: FIELDTRIP at NIC.SURFNET.NL > > This probably has to do with EEGLAB being above Fieldtrip in your Matlab > path with EEGLAB's version of topoplot being called instead of Fieldtrip's. > You need to put Fieldtrip on top and make sure it stays there because EEGLAB > moves itself to the top each time it's called. Perhaps you should add a line > modifying the path to your script. > > Best, > > Vladimir > > -----Original Message----- > From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf > Of Frederic Roux > Sent: Friday, January 04, 2008 10:04 AM > To: FIELDTRIP at NIC.SURFNET.NL > Subject: [FIELDTRIP] topoplot error > > Hi everybody, > > I've been trying to use the topoplotTFR function but constantly receive the > error message 'Odd number of arguments?'. > Has anyone ever encountered that problem? > > I used the following commands: > > cfg = []; > cfg.layout = prepare_layout(data); > cfg.baselinetype = 'relative'; > cfg.xparam = 'time'; > cfg.yparam = 'freq'; > cfg.zparam = 'powspctrm'; > cfg.xlim = 'maxmin'; > cfg.ylim = 'maxmin'; > cfg.zlim = 'maxmin'; > > topoplotTFR(cfg,data) > > topoplotTFR calls topoplotER which again calls topoplot. > > The problem seems to be related to the following lines of code: > > in topoplotER @ line 364: > > topoplot(tmpcfg,chanX,chanY,datavector,chanLabels); > > and in topoplot @ line 957: > > if nargs> 2 > if ~(round(nargs/2) == nargs/2) > error('Odd number of input arguments??') > end > > I don't get any further. Can anyone help me, please? > > Regards, > > Frederic > _________________________________________________________________ > Importieren Sie ganz einfach Ihre E-Mail Adressen in den Messenger! > http://messenger.live.de/community/neuekontakte_adressimport.html > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. _________________________________________________________________ Windows Live Fotogalerie: So einfach organisieren Sie Ihre Fotos! http://get.live.com/photogallery/overview ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Sun Jan 6 18:40:59 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Sun, 6 Jan 2008 17:40:59 +0000 Subject: topoplot error In-Reply-To: Message-ID: Hi everyone, Does anyone know where to find the standardised montreal " average" mri in ctf format? I see that there are some average mri's in the spm toolbox(spm2\canonical) but these are .mnc files in spm coordinates. I have two subjects without mris and have found that the localisation is much more accurate if i use another subjects ctf mri than if i use the mri's provided with spm (to produce the headmodel and for source interpolation). The MEG scanner used was a ctf machine so i thought that using a ctf mri average would give the best result. I have been careful to ensure that the coordinate system used both for the headmodel formation and for interpolation has been defined appropriately for the type of mri being used (eg cfg.coordinates = 'ctf'; or ='spm'). Any thoughts would be much appreciated. Also i cant seem to find the function beamformer_pcc.m which is called when using sourceanalysis.m, cfg. method ='pcc'; knowledge of its whereabouts would be greatly received! best wishes tom gilbertson ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at FCDONDERS.RU.NL Mon Jan 7 15:03:55 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Mon, 7 Jan 2008 15:03:55 +0100 Subject: automatic update wiki/ftp disabled for a few days Message-ID: Dear all The automatic updating of the documentation on the fieldtrip website and the automatic updating of the fieldtrip code on /home/common (internal to FC Donders) and on the external ftp server are TEMPORARY DISABLED. This is necessary due to migrating the cvs repository to a new server and to implementing new release management scripts. The automatic updating will be up again as soon as possible, but it might take a few days. best regards, Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Mon Jan 7 15:18:22 2008 From: jan.schoffelen at FCDONDERS.RU.NL (Jan Mathijs Schoffelen) Date: Mon, 7 Jan 2008 15:18:22 +0100 Subject: forward modelling without anatomical info Message-ID: Dear Tom, As far as I know, the 'ctf-format' standardised mri does not exist. Creating an anatomically accurate forward model, while the actual anatomical information is lacking, is tricky business. What kind of forward model do you actully use? The multiple sphere model based on the ctf-shape files? If so, you could try to come up with an 'averaged' multipsphere model (thus bypassing the need for a standardised MRI), although this is not trial. However, I wonder whether this would yield better results than just taking an 'averaged' single sphere model. I wonder what you mean by your statement that localisation is much more accurate when using ctf-formatted MRIs. Do you know where your activation should end up then? The function beamformer_pcc is as far as I know still in its experimental stage and only for use within the FC Donders Centre. However, beamformer_dics shares most of the functionality, but obviously not all of it. Regards, Jan-Mathijs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 7 16:59:09 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 7 Jan 2008 15:59:09 +0000 Subject: forward modelling without anatomical info In-Reply-To: <3c02ed1911ff6.478242be@ru.nl> Message-ID: dear jan mathijs, i have been generating a single shell model from the segmented mri. the localisation proceedure is for a tactile stimulus so when i see a contralateral S1 response i have some reason to belive that the algorithm is making some sense. I get the impression from your reply that there is no easy way to get around this lack of anatomical data and that i may just have to go and get the mri's done for this couple of subjects! thanks for your thoughts on this, tom. On 1/7/08, Jan Mathijs Schoffelen wrote: > > Dear Tom, > > As far as I know, the 'ctf-format' standardised mri does not exist. > Creating an anatomically accurate forward model, while the actual anatomical > information is lacking, is tricky business. What kind of forward model do > you actully use? The multiple sphere model based on the ctf-shape files? If > so, you could try to come up with an 'averaged' multipsphere model (thus > bypassing the need for a standardised MRI), although this is not > trial. However, I wonder whether this would yield better results than just > taking an 'averaged' single sphere model. > I wonder what you mean by your statement that localisation is much more > accurate when using ctf-formatted MRIs. Do you know where your activation > should end up then? > > The function beamformer_pcc is as far as I know still in its experimental > stage and only for use within the FC Donders Centre. However, > beamformer_dics shares most of the functionality, but obviously not all of > it. > > Regards, > > Jan-Mathijs > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From weisz at LYON.INSERM.FR Mon Jan 7 17:11:57 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Mon, 7 Jan 2008 17:11:57 +0100 Subject: forward modelling without anatomical info In-Reply-To: Message-ID: hi, on the fieldtrip FTP there is a standard BEM (standard_vol.matr). i think you could use the skull part as pseudo-headshape: vol.bnd(1).pnt this also fits with the standard MRI. there may be some transformation you have to apply to these "headshape" points so that the head fits nicely into the sensor array. then you can use prepare_localspheres and specify a lower number in cfg.spheremesh (perhaps something like 50; default is 4000). this may yield a crude but satisfactory solution if you are unable to get MRIs or individual headshapes. i'd be interested in the outcome. all the best, nathan On 07.01.2008, at 16:59, thomas gilbertson wrote: > dear jan mathijs, > > i have been generating a single shell model from the segmented mri. > the localisation proceedure is for a tactile stimulus so when i see > a contralateral S1 response i have some reason to belive that the > algorithm is making some sense. I get the impression from your reply > that there is no easy way to get around this lack of anatomical data > and that i may just have to go and get the mri's done for this > couple of subjects! > > thanks for your thoughts on this, > > tom. > > > On 1/7/08, Jan Mathijs Schoffelen > wrote: > Dear Tom, > > As far as I know, the 'ctf-format' standardised mri does not exist. > Creating an anatomically accurate forward model, while the actual > anatomical information is lacking, is tricky business. What kind of > forward model do you actully use? The multiple sphere model based on > the ctf-shape files? If so, you could try to come up with an > 'averaged' multipsphere model (thus bypassing the need for a > standardised MRI), although this is not trial. However, I wonder > whether this would yield better results than just taking an > 'averaged' single sphere model. > I wonder what you mean by your statement that localisation is much > more accurate when using ctf-formatted MRIs. Do you know where your > activation should end up then? > > The function beamformer_pcc is as far as I know still in its > experimental stage and only for use within the FC Donders Centre. > However, beamformer_dics shares most of the functionality, but > obviously not all of it. > > Regards, > > Jan-Mathijs > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 7 17:17:11 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 7 Jan 2008 16:17:11 +0000 Subject: forward modelling without anatomical info In-Reply-To: <99082A9C-EC3C-4EDE-884F-5D3D5AD3FADB@lyon.inserm.fr> Message-ID: thanks nathan, ill give it a try! tom On 1/7/08, Nathan Weisz wrote: > > hi, > > > on the fieldtrip FTP there is a standard BEM (standard_vol.matr). i think > you could use the skull part as pseudo-headshape: > vol.bnd(1).pnt > > > this also fits with the standard MRI. > > > there may be some transformation you have to apply to these "headshape" > points so that the head fits nicely into the sensor array. > then you can use prepare_localspheres and specify a lower number in > cfg.spheremesh (perhaps something like 50; default is 4000). > > > this may yield a crude but satisfactory solution if you are unable to get > MRIs or individual headshapes. > > > i'd be interested in the outcome. > > > all the best, > nathan > > > > > > > > On 07.01.2008, at 16:59, thomas gilbertson wrote: > > dear jan mathijs, > > i have been generating a single shell model from the segmented mri. > the localisation proceedure is for a tactile stimulus so when i see a > contralateral S1 response i have some reason to belive that the algorithm is > making some sense. I get the impression from your reply that there is no > easy way to get around this lack of anatomical data and that i may just have > to go and get the mri's done for this couple of subjects! > > thanks for your thoughts on this, > > tom. > > > On 1/7/08, Jan Mathijs Schoffelen wrote: > > > > Dear Tom, > > > > As far as I know, the 'ctf-format' standardised mri does not exist. > > Creating an anatomically accurate forward model, while the actual anatomical > > information is lacking, is tricky business. What kind of forward model do > > you actully use? The multiple sphere model based on the ctf-shape files? If > > so, you could try to come up with an 'averaged' multipsphere model (thus > > bypassing the need for a standardised MRI), although this is not > > trial. However, I wonder whether this would yield better results than just > > taking an 'averaged' single sphere model. > > I wonder what you mean by your statement that localisation is much more > > accurate when using ctf-formatted MRIs. Do you know where your activation > > should end up then? > > > > The function beamformer_pcc is as far as I know still in its > > experimental stage and only for use within the FC Donders Centre. However, > > beamformer_dics shares most of the functionality, but obviously not all of > > it. > > > > Regards, > > > > Jan-Mathijs > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > > the FieldTrip toolbox, to share experiences and to discuss new ideas for > > MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and > EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > > > -------------------------------- > *Dr. Nathan Weisz* > > INSERM - Unité 821 > *Dynamique cérébrale et cognition* > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > *Tel*: ++33 - (0)4 - 7213 8915 > *Email: *weisz at lyon.inserm.fr > *Chat-AV*: nathanweisz at mac.com > *Homepage*: http://web.mac.com/nathanweisz > *Neurotree*: http://neurotree.org/neurotree/tree.php?pid=8692 > > > *Please avoid sending me Word or PowerPoint attachments.** > **See http://www.gnu.org/philosophy/no-word-attachments.html* > > > > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of the > FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and > EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From linpe at NINDS.NIH.GOV Tue Jan 8 21:34:37 2008 From: linpe at NINDS.NIH.GOV (Peter Lin) Date: Tue, 8 Jan 2008 21:34:37 +0100 Subject: DICS pcc error Message-ID: I have a question. We are trying to process data to use DICS using fieldtrip at NIH using CTF system. In using the pcc script for preprocessing, we get this error: ??? Error using ==> ctf2grad at 68 unexpected number of coils in MEG channel Error in ==> read_header at 305 hdr.grad = ctf2grad(orig); Error in ==> read_fcdc_header at 37 [hdr] = read_header(varargin{:}); Error in ==> preprocessing at 291 hdr = read_fcdc_header(cfg.headerfile); In looking at the MEG channels, there are just the standard MEG ref channels, EEG channels (6) and MEG channels. Not sure where the program is going wrong. this is 20070204 version of fieldtrip. Thanks, peter Lin NINDS/NIH ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Wed Jan 9 08:23:45 2008 From: jan.schoffelen at FCDONDERS.RU.NL (Jan Mathijs Schoffelen) Date: Wed, 9 Jan 2008 08:23:45 +0100 Subject: DICS pcc error Message-ID: Dear Peter, First, I would advice you to update your fieldtrip-version to a more recent one. Apparently something is going wrong in read_fcdc_header, and not in the beamformer-algorithm. It seems as if the gradiometer-information in your headerfile (which is the .res4-file in the ctf-dataset) does not contain the information which fieldTrip expects. You might have to look into that. One likely cause is the absence of head-position information in your dataset. As a consequence the gradiometer-info in your .res4-file will only be specified in dewar-coordinates, which makes the dataset useless for source localisation anyway, because the MEG-data cannot be aligned to an anatomical coordinate system. Yours, Jan-Mathijs ----- Original Message ----- From: Peter Lin Date: Tuesday, January 8, 2008 9:34 pm Subject: [FIELDTRIP] DICS pcc error > I have a question. We are trying to process data to use DICS using > fieldtrip at NIH using CTF system. In using the pcc script for > preprocessing, we get this error: > > ??? Error using ==> ctf2grad at 68 > unexpected number of coils in MEG channel > > Error in ==> read_header at 305 > hdr.grad = ctf2grad(orig); > > Error in ==> read_fcdc_header at 37 > [hdr] = read_header(varargin{:}); > > Error in ==> preprocessing at 291 > hdr = read_fcdc_header(cfg.headerfile); > > In looking at the MEG channels, there are just the standard MEG ref > channels, EEG channels (6) and MEG channels. Not sure where the > program is > going wrong. this is 20070204 version of fieldtrip. > > Thanks, > > peter Lin > NINDS/NIH > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jedmeltzer at YAHOO.COM Thu Jan 10 00:34:34 2008 From: jedmeltzer at YAHOO.COM (Jed Meltzer) Date: Wed, 9 Jan 2008 15:34:34 -0800 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: Message-ID: The error may possibly have something to with the fact that our CTF system at NIH is missing two channels (due to broken squids) - maybe the software expects the full 275 to be there. I don't know though, as I have successfully imported 273-channel data from that system into fieldtrip, although I had other problems past the import stage that I have not gotten around to troubleshooting... -Jed --- FIELDTRIP automatic digest system wrote: --------------------------------- FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Service Desk : LISTSERV Archives FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Table of contents: DICS pcc error DICS pcc error Re: DICS pcc error (01/09) From: Jan Mathijs Schoffelen --------------------------------- Browse the FIELDTRIP online archives.> Date: Wed, 9 Jan 2008 08:23:45 +0100 > From: Jan Mathijs Schoffelen > > Subject: Re: DICS pcc error > > Dear Peter, > > First, I would advice you to update your > fieldtrip-version to a more recent one. > > Apparently something is going wrong in > read_fcdc_header, and not in the > beamformer-algorithm. It seems as if the > gradiometer-information in your headerfile (which is > the .res4-file in the ctf-dataset) does not contain > the information which fieldTrip expects. You might > have to look into that. One likely cause is the > absence of head-position information in your > dataset. As a consequence the gradiometer-info in > your .res4-file will only be specified in > dewar-coordinates, which makes the dataset useless > for source localisation anyway, because the MEG-data > cannot be aligned to an anatomical coordinate > system. > > Yours, > > Jan-Mathijs > > > ----- Original Message ----- > From: Peter Lin > Date: Tuesday, January 8, 2008 9:34 pm > Subject: [FIELDTRIP] DICS pcc error > > > I have a question. We are trying to process data > to use DICS using > > fieldtrip at NIH using CTF system. In using the > pcc script for > > preprocessing, we get this error: > > > > ??? Error using ==> ctf2grad at 68 > > unexpected number of coils in MEG channel > > > > Error in ==> read_header at 305 > > hdr.grad = ctf2grad(orig); > > > > Error in ==> read_fcdc_header at 37 > > [hdr] = read_header(varargin{:}); > > > > Error in ==> preprocessing at 291 > > hdr = read_fcdc_header(cfg.headerfile); > > > > In looking at the MEG channels, there are just the > standard MEG ref > > channels, EEG channels (6) and MEG channels. Not > sure where the > > program is > > going wrong. this is 20070204 version of > fieldtrip. > > > > Thanks, > > > > peter Lin > > NINDS/NIH > > > > ---------------------------------- > > The aim of this list is to facilitate the > discussion between users > > of the FieldTrip toolbox, to share experiences > and to discuss new > > ideas for MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html > and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > The aim of this list is to facilitate the discussion > between users of the FieldTrip toolbox, to share > experiences and to discuss new ideas for MEG and EEG > analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From linpe at NINDS.NIH.GOV Thu Jan 10 01:47:44 2008 From: linpe at NINDS.NIH.GOV (Lin, Peter (NIH/NINDS) [E]) Date: Wed, 9 Jan 2008 19:47:44 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Message-ID: HI Jed, thanks. Does Tom know? Peter -----Original Message----- From: Jed Meltzer [mailto:jedmeltzer at YAHOO.COM] Sent: Wed 1/9/2008 6:34 PM To: FIELDTRIP at NIC.SURFNET.NL Subject: Re: [FIELDTRIP] FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) The error may possibly have something to with the fact that our CTF system at NIH is missing two channels (due to broken squids) - maybe the software expects the full 275 to be there. I don't know though, as I have successfully imported 273-channel data from that system into fieldtrip, although I had other problems past the import stage that I have not gotten around to troubleshooting... -Jed --- FIELDTRIP automatic digest system wrote: --------------------------------- FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Service Desk : LISTSERV Archives FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) Table of contents: DICS pcc error DICS pcc error Re: DICS pcc error (01/09) From: Jan Mathijs Schoffelen --------------------------------- Browse the FIELDTRIP online archives.> Date: Wed, 9 Jan 2008 08:23:45 +0100 > From: Jan Mathijs Schoffelen > > Subject: Re: DICS pcc error > > Dear Peter, > > First, I would advice you to update your > fieldtrip-version to a more recent one. > > Apparently something is going wrong in > read_fcdc_header, and not in the > beamformer-algorithm. It seems as if the > gradiometer-information in your headerfile (which is > the .res4-file in the ctf-dataset) does not contain > the information which fieldTrip expects. You might > have to look into that. One likely cause is the > absence of head-position information in your > dataset. As a consequence the gradiometer-info in > your .res4-file will only be specified in > dewar-coordinates, which makes the dataset useless > for source localisation anyway, because the MEG-data > cannot be aligned to an anatomical coordinate > system. > > Yours, > > Jan-Mathijs > > > ----- Original Message ----- > From: Peter Lin > Date: Tuesday, January 8, 2008 9:34 pm > Subject: [FIELDTRIP] DICS pcc error > > > I have a question. We are trying to process data > to use DICS using > > fieldtrip at NIH using CTF system. In using the > pcc script for > > preprocessing, we get this error: > > > > ??? Error using ==> ctf2grad at 68 > > unexpected number of coils in MEG channel > > > > Error in ==> read_header at 305 > > hdr.grad = ctf2grad(orig); > > > > Error in ==> read_fcdc_header at 37 > > [hdr] = read_header(varargin{:}); > > > > Error in ==> preprocessing at 291 > > hdr = read_fcdc_header(cfg.headerfile); > > > > In looking at the MEG channels, there are just the > standard MEG ref > > channels, EEG channels (6) and MEG channels. Not > sure where the > > program is > > going wrong. this is 20070204 version of > fieldtrip. > > > > Thanks, > > > > peter Lin > > NINDS/NIH > > > > ---------------------------------- > > The aim of this list is to facilitate the > discussion between users > > of the FieldTrip toolbox, to share experiences > and to discuss new > > ideas for MEG and EEG analysis. See also > > http://listserv.surfnet.nl/archives/fieldtrip.html > and > > http://www.ru.nl/fcdonders/fieldtrip. > > > > ---------------------------------- > The aim of this list is to facilitate the discussion > between users of the FieldTrip toolbox, to share > experiences and to discuss new ideas for MEG and EEG > analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. > ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From venug001 at BAMA.UA.EDU Thu Jan 10 20:00:13 2008 From: venug001 at BAMA.UA.EDU (Gopa Venugopalan) Date: Thu, 10 Jan 2008 20:00:13 +0100 Subject: Importing 4-D datasets Message-ID: Hello, I wish everyone a productive New Year! Does anyone have any experience importing 4-D data? I have 146 channels of MEG and 25 channels of EEG that I am importing. That data has been preprocessed and grand averaged using proprretary 4-D software. All I now need is to import the ASCII files containing the averaged data. Thanking everyone in advance for any and all assistance. Warm Regards gopa ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 10 23:06:31 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 10 Jan 2008 17:06:31 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <265F73253EB9234F81FF4EC2DCB2AD1CA8B25B@nihcesmlbx2.nih.gov> Message-ID: On Wed, 2008-01-09 at 19:47 -0500, Lin, Peter (NIH/NINDS) [E] wrote: > HI Jed, > > thanks. Does Tom know? Tom is working on it. All I know at this point is that I tried it with Octave 3.0 (which just came out recently) and it works in Octave. Fieldtrip comes up with way too many markers in the stim channel for some reason under recent matlab, but it's OK in Octave. (Ha!) Dr. Tom -- My brother, when you have a virtue, and it is your own virtue, you have it in common with no one. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 09:41:26 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 09:41:26 +0100 Subject: Importing 4-D datasets In-Reply-To: Message-ID: hi gopa, the 4-D software should come with a command line tool called pdf2set. files exported with this command can be read in a standard fieldtrip manner. yet this assumes you want to read trials. you already averaged the data and probably want a timelockanalysis structure. perhaps it is the least pain to read the data (exported via pdf2set) with eeglab. then do dummy = eeglab2fieldtrip(EEG, 'timelockanalysis'). best wishes, nathan On 10.01.2008, at 20:00, Gopa Venugopalan wrote: > Hello, I wish everyone a productive New Year! Does anyone have any > experience importing 4-D data? I have 146 channels of MEG and 25 > channels > of EEG that I am importing. That data has been preprocessed and grand > averaged using proprretary 4-D software. All I now need is to import > the > ASCII files containing the averaged data. Thanking everyone in > advance for > any and all assistance. > Warm Regards > gopa > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html > and http://www.ru.nl/fcdonders/fieldtrip. -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ewingerd at SCIENCE.UVA.NL Fri Jan 11 10:22:04 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Fri, 11 Jan 2008 10:22:04 +0100 Subject: non-uniform trial length In-Reply-To: Message-ID: Hi Fieldtrippers, Recently, I've begun trying to use FieldTrip to analyse LFP traces from rats implanted with tetrode arrays. I would like to look at changes in power over a range of frequencies in relation to behavioural events. So, likely, I will use freqanalysis_tfr as a method for this. Since I'm interested in /changes/ in power from baseline, I would probably have to use freqbaseline with the 'relative change' method. Looking at the code for this function, it seems that it expects a baseline period somewhere before the event of interest, for each trial. In my research, the rats are freely moving around a cage, so there is no fixed interval between the end of the baseline period and the start of the peri-event interval of interest. As I see it, the baselinefreq function requires the baseline period to be a fixed part of any trial. Is there already a way to pass a full, pre-made freq.powspctrm (from the baseline periods) to the freqbaseline function? I noticed that the subfunction TFrelchange indexes into the event intervals. I guess I could rewrite that part as a private function to accept such a baseline freq.powspctrm, but I'm wondering if such a tool already exists. Alternatively, I guess I could create a 'trialfun' function that stitches the baseline and event part of the data together on a trial-by-trial basis, resulting in a uniform trial length. I would greatly appreciate any feedback on whether to go with A or B, or which solution would work the best in the FieldTrip framework, thanks, marijn -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 11:54:06 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 11:54:06 +0100 Subject: non-uniform trial length In-Reply-To: <4787353C.8020605@science.uva.nl> Message-ID: Dear Marjin, i don't know if it's useful for you but i adapted the freqbaseline function (attached). using cfg.otherbl you can pass on a freqanalysis another structure which will then be used to baseline-correct your freqanalysis structure of interest (see the help). if you decide to test it, I would be interested if it works for you. also I have a colleague who would also like to analyse rat data with Fieldtrip. I'd be interested in your opinion if its well-suited. btw, for time-frequency stuff i have good experiences with freqanalysis_wltconvol. all the best, nathan On 11.01.2008, at 10:22, Marijn van Wingerden wrote: > Marijn van Wingerden -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: freqbaseline.m Type: application/octet-stream Size: 13658 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: From venug001 at BAMA.UA.EDU Fri Jan 11 13:03:07 2008 From: venug001 at BAMA.UA.EDU (Gopakumar Venugopalan) Date: Fri, 11 Jan 2008 06:03:07 -0600 Subject: Importing 4-D datasets In-Reply-To: <159CABC0-0AB1-4572-A4E7-38160064A0AC@lyon.inserm.fr> Message-ID: Thanks Nathan. I did install pdf2set. The 4-D data is not discrimnating between EEG and MEG as the EEG electrodes were not digitized. So it is exported as one large 171 channel file. Yes, I guess timelock structure is where I want to start from then move to Topographic analysis, areal analysis and dipole fitting. Would you happen to know how many files pdf2set will export and with what extensions? I have ascii files of the data sets, I saw an .xyz file (cartesian coordinates) and an elp file (electrode position) file. Can I use the last two files in conjunction with the standard 10-20 electrode system position to read in the EEG data? Thanks, once again, in advance. Warm regards gopa Quoting Nathan Weisz : > hi gopa, > > the 4-D software should come with a command line tool called pdf2set. > > files exported with this command can be read in a standard fieldtrip > > manner. yet this assumes you want to read trials. > you already averaged the data and probably want a timelockanalysis > structure. perhaps it is the least pain to read the data (exported > via > pdf2set) with eeglab. then do > dummy = eeglab2fieldtrip(EEG, 'timelockanalysis'). > > best wishes, > nathan > > > > > > > On 10.01.2008, at 20:00, Gopa Venugopalan wrote: > > > Hello, I wish everyone a productive New Year! Does anyone have any > > experience importing 4-D data? I have 146 channels of MEG and 25 > > channels > > of EEG that I am importing. That data has been preprocessed and > grand > > averaged using proprretary 4-D software. All I now need is to > import > > the > > ASCII files containing the averaged data. Thanking everyone in > > advance for > > any and all assistance. > > Warm Regards > > gopa > > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users > > > of the FieldTrip toolbox, to share experiences and to discuss new > > > ideas for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html > > and http://www.ru.nl/fcdonders/fieldtrip. > > -------------------------------- > Dr. Nathan Weisz > > INSERM - Unité 821 > Dynamique cérébrale et cognition > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > Tel: ++33 - (0)4 - 7213 8915 > Email: weisz at lyon.inserm.fr > Chat-AV: nathanweisz at mac.com > Homepage: http://web.mac.com/nathanweisz > Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 > > Please avoid sending me Word or PowerPoint attachments. > See http://www.gnu.org/philosophy/no-word-attachments.html > > > > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From l.frei at PSY.GLA.AC.UK Fri Jan 11 13:47:44 2008 From: l.frei at PSY.GLA.AC.UK (Luisa Frei) Date: Fri, 11 Jan 2008 12:47:44 +0000 Subject: missing files, reshape Message-ID: Hi, I’m currently doing the fieldtrip tutorials and I have encountered 2 problems: 1. Some of the pre-calculated data files are missing. That’s especially complicating for the Parametric and non-parametric statistics tutorial, because the whole tutorial relies on that file: GA_ERF_orig.mat I also don't have the files for the cluster-based permutation test tutorial. Does anyone know, where I can find them? 2. In the cluster-based permutation test tutorial I get an error with the reshape function at the end of 2.1.2.1. Here’s the error text: >> [stat] = timelockstatistics(cfg, FICdata, FCdata); ??? Error using ==> reshape To RESHAPE the number of elements must not change. Error in ==> fieldtrip-20080107/private/ prepare_timefreq_data>forcedimord at 674 output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); Error in ==> timelockstatistics at 107 [stat] = statistics_wrapper(cfg, varargin{:}); I didn’t manage to fix this. Has anyone else encountered this and knows how to solve it? Any help would be greatly appreciated. Thanks, Luisa ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From weisz at LYON.INSERM.FR Fri Jan 11 14:28:40 2008 From: weisz at LYON.INSERM.FR (Nathan Weisz) Date: Fri, 11 Jan 2008 14:28:40 +0100 Subject: Importing 4-D datasets In-Reply-To: <1200052987.47875afbab7fa@bamamail.ua.edu> Message-ID: hi gopa, > between EEG and MEG as the EEG electrodes were not digitized. So it is > exported as one large 171 channel file. using: hdr=read_fcdc_header('yourm4dfile.m4d') and then: hdr.grad.label you shoud see which sensors / electrodes correspond to which row in your data matrix. > Would you happen to know how many files pdf2set will export and with > what extensions? I have ascii files of the data sets, I saw an .xyz > file (cartesian coordinates) and an elp file (electrode position) > file. next to the data and the xyz you should have the header-file (m4d). if you want to do dipole fits you also need the headshape file (hs_file; read it via read_bti_hs). [just a side note: if it's a simple dipole fit you want then the 4d software should be good enough, so why go through the pain of exporting and importing data] > Can I use the last two files in conjunction with the standard 10-20 > electrode system position to read in the EEG data? I am not quite sure if i understand you correctly. but many fieldtrip functions (e.g. definetrial, timelockanalysis, freqanalysis) allow you to specifiy the channels you want, e.g.: cfg.channel={'MEG'}; you may try: cfg.channel={'EEG'}; and see if it only applies the relevant functions to the EEG channels (i haven't done this myself). if not you can specify all labels by hand or add something to the channelselction.m function that suits your needs. regarding electrode positions on the fieldtrip-FTP (standard_BEM) there are standard 10-20 and 10-05 elc files. you can read those files using read_fcdc_elec. hope this helps, nathan -------------------------------- Dr. Nathan Weisz INSERM - Unité 821 Dynamique cérébrale et cognition Centre Hospitalier Le Vinatier, Bâtiment 452 95 Boulevard Pinel 69500 Bron, France Tel: ++33 - (0)4 - 7213 8915 Email: weisz at lyon.inserm.fr Chat-AV: nathanweisz at mac.com Homepage: http://web.mac.com/nathanweisz Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692 Please avoid sending me Word or PowerPoint attachments. See http://www.gnu.org/philosophy/no-word-attachments.html ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ewingerd at SCIENCE.UVA.NL Tue Jan 15 09:51:03 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Tue, 15 Jan 2008 09:51:03 +0100 Subject: non-uniform trial length In-Reply-To: Message-ID: Dear Nathan, I think that is exactly the fix I was looking for. I'll let you know what comes out of testing the .m file thanks a lot, all the best, marijn Nathan Weisz wrote: > Dear Marjin, > > i don't know if it's useful for you but i adapted the freqbaseline > function (attached). > using cfg.otherbl > you can pass on a freqanalysis another structure which will then be > used to baseline-correct your freqanalysis structure of interest (see > the help). > > if you decide to test it, I would be interested if it works for you. > also I have a colleague who would also like to analyse rat data with > Fieldtrip. I'd be interested in your opinion if its well-suited. > > btw, for time-frequency stuff i have good experiences > with freqanalysis_wltconvol. > > all the best, > nathan > > > > On 11.01.2008, at 10:22, Marijn van Wingerden wrote: > >> Marijn van Wingerden > > -------------------------------- > *Dr. Nathan Weisz* > > INSERM - Unité 821 > /Dynamique cérébrale et cognition/ > Centre Hospitalier Le Vinatier, Bâtiment 452 > 95 Boulevard Pinel > 69500 Bron, France > > *Tel*: ++33 - (0)4 - 7213 8915 > *Email: *weisz at lyon.inserm.fr > *Chat-AV*: nathanweisz at mac.com > *Homepage*: http://web.mac.com/nathanweisz > *Neurotree*: http://neurotree.org/neurotree/tree.php?pid=8692 > > /Please avoid sending me Word or PowerPoint attachments.// > //See http://www.gnu.org/philosophy/no-word-attachments.html/ > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > > > ------------------------------------------------------------------------ > > > ---------------------------------- > > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. > > http://listserv.surfnet.nl/archives/fieldtrip.html > > http://www.ru.nl/fcdonders/fieldtrip/ > -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From jan.schoffelen at FCDONDERS.RU.NL Tue Jan 15 12:20:25 2008 From: jan.schoffelen at FCDONDERS.RU.NL (jan-mathijs schoffelen) Date: Tue, 15 Jan 2008 12:20:25 +0100 Subject: missing files, reshape In-Reply-To: <4FE97248-C603-452A-AA08-ED56D7E18EBB@PSY.GLA.AC.UK> Message-ID: Dear Luisa, I put the missing GA_ERF_orig mat-file on the ftp-server. You can access it at: ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail address). It's in the directory /pub/temporary. As far as I can see, the second problem could be related to the fact that you do not have the correct data. It might be solved when you use the data in the mat-file. Yours, Jan-Mathijs On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > Hi, > I’m currently doing the fieldtrip tutorials and I have encountered > 2 problems: > > 1. Some of the pre-calculated data files are missing. That’s > especially complicating for the Parametric and non-parametric > statistics tutorial, because the whole tutorial relies on that > file: GA_ERF_orig.mat > I also don't have the files for the cluster-based permutation test > tutorial. Does anyone know, where I can find them? > > 2. In the cluster-based permutation test tutorial I get an error > with the reshape function at the end of 2.1.2.1. Here’s the error > text: > > >> [stat] = timelockstatistics(cfg, FICdata, FCdata); > ??? Error using ==> reshape > To RESHAPE the number of elements must not change. > > Error in ==> fieldtrip-20080107/private/ > prepare_timefreq_data>forcedimord at 674 > output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); > > Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 > [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); > > Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 > [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); > > Error in ==> timelockstatistics at 107 > [stat] = statistics_wrapper(cfg, varargin{:}); > > > I didn’t manage to fix this. Has anyone else encountered this and > knows how to solve it? Any help would be greatly appreciated. > > > Thanks, > Luisa > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From ingrid.nieuwenhuis at FCDONDERS.RU.NL Tue Jan 15 12:41:59 2008 From: ingrid.nieuwenhuis at FCDONDERS.RU.NL (Ingrid Nieuwenhuis) Date: Tue, 15 Jan 2008 12:41:59 +0100 Subject: missing files, reshape In-Reply-To: Message-ID: Dear Louisa, The second problem might also be related to the dof field. If there's a dof field in the data, removing it might be a fix. Best Ingrid -----Original Message----- From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf Of jan-mathijs schoffelen Sent: Tuesday, January 15, 2008 12:20 PM To: FIELDTRIP at NIC.SURFNET.NL Subject: Re: [FIELDTRIP] missing files, reshape Dear Luisa, I put the missing GA_ERF_orig mat-file on the ftp-server. You can access it at: ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail address). It's in the directory /pub/temporary. As far as I can see, the second problem could be related to the fact that you do not have the correct data. It might be solved when you use the data in the mat-file. Yours, Jan-Mathijs On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > Hi, > I'm currently doing the fieldtrip tutorials and I have encountered > 2 problems: > > 1. Some of the pre-calculated data files are missing. That's > especially complicating for the Parametric and non-parametric > statistics tutorial, because the whole tutorial relies on that > file: GA_ERF_orig.mat > I also don't have the files for the cluster-based permutation test > tutorial. Does anyone know, where I can find them? > > 2. In the cluster-based permutation test tutorial I get an error > with the reshape function at the end of 2.1.2.1. Here's the error > text: > > >> [stat] = timelockstatistics(cfg, FICdata, FCdata); > ??? Error using ==> reshape > To RESHAPE the number of elements must not change. > > Error in ==> fieldtrip-20080107/private/ > prepare_timefreq_data>forcedimord at 674 > output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); > > Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 > [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); > > Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 > [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); > > Error in ==> timelockstatistics at 107 > [stat] = statistics_wrapper(cfg, varargin{:}); > > > I didn't manage to fix this. Has anyone else encountered this and > knows how to solve it? Any help would be greatly appreciated. > > > Thanks, > Luisa > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From r.oostenveld at FCDONDERS.RU.NL Wed Jan 16 09:11:52 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Wed, 16 Jan 2008 09:11:52 +0100 Subject: Importing 4-D datasets In-Reply-To: <02BA90E7-72D6-4BA1-9DF6-45C678A1F60D@lyon.inserm.fr> Message-ID: Hi Gopa Regarding combined EEG and MEG data, of course we also have this often with our CTF MEG system. The sensor locations stored in a preprocessed combined EEG+MEG data structure at the moment can not contain mixed EEG and MEG positions. Usually you will want to do source reconstruction with the MEG channels. In that case the data (after preprocessing, timelockanalysis or freqanalysis or componentanalysis) should have the field data.grad. If you want to do sourceanalysis/dipolefitting on the EEG channels, you should remove the data.grad field and instead insert a field into the data with the name data.elec. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes _magnetometers_or_gradiometers_described for fuether details. On 11 Jan 2008, at 14:28, Nathan Weisz wrote: > you may try: > cfg.channel={'EEG'}; > and see if it only applies the relevant functions to the EEG > channels (i haven't done this myself). if not you can specify all > labels by hand or add something to the channelselction.m function > that suits your needs. Usefull here might also be to specify cfg.channel='gui' and a graphical interface will pop up allowing you to select channels with the mouse. best regards, Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From venug001 at BAMA.UA.EDU Wed Jan 16 13:36:27 2008 From: venug001 at BAMA.UA.EDU (Gopakumar Venugopalan) Date: Wed, 16 Jan 2008 06:36:27 -0600 Subject: Importing 4-D datasets In-Reply-To: Message-ID: Greeting Robert, thanks for the freeware and assistance. Nathan has been very,very helpful from the outset. I am truly grateful to all out there best regards gopa Quoting Robert Oostenveld : > Hi Gopa > > Regarding combined EEG and MEG data, of course we also have this > often with our CTF MEG system. The sensor locations stored in a > preprocessed combined EEG+MEG data structure at the moment can not > contain mixed EEG and MEG positions. Usually you will want to do > source reconstruction with the MEG channels. In that case the data > (after preprocessing, timelockanalysis or freqanalysis or > componentanalysis) should have the field data.grad. If you want to do > > sourceanalysis/dipolefitting on the EEG channels, you should remove > > the data.grad field and instead insert a field into the data with the > > name data.elec. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? > > id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes > > _magnetometers_or_gradiometers_described for fuether details. > > On 11 Jan 2008, at 14:28, Nathan Weisz wrote: > > you may try: > > cfg.channel={'EEG'}; > > and see if it only applies the relevant functions to the EEG > > channels (i haven't done this myself). if not you can specify all > > > labels by hand or add something to the channelselction.m function > > > that suits your needs. > > Usefull here might also be to specify cfg.channel='gui' and a > graphical interface will pop up allowing you to select channels with > > the mouse. > > best regards, > Robert > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From r.oostenveld at FCDONDERS.RU.NL Wed Jan 16 17:32:14 2008 From: r.oostenveld at FCDONDERS.RU.NL (Robert Oostenveld) Date: Wed, 16 Jan 2008 17:32:14 +0100 Subject: topoplot error In-Reply-To: Message-ID: Hi Thomas, On 6 Jan 2008, at 18:40, thomas gilbertson wrote: > Does anyone know where to find the standardised montreal " average" > mri in ctf format? Maybe a bit late, but I'll give a short reply anyway. If you read the canonical SPM MRI using mri=read_fcdc_mri(..), it will have mri.transform describing the homogenous coordinate transformation matrix that converts from voxel indices into SPM coordinates. In volumenormalise around line 153, there is a homogenous transformation matrix that converts an MRI from CTF head coordinates (linked to NAS/LPA/RPA) into SPM coordinates (linked to AC/PC). If you multiply the inverse of that 4x4 matrix with mri.transform that you obtained after reading the SPM canonical MRI, you should get the canonical MRI reasonably aligned with the CTF headcoordinate system. See http://www2.ru.nl/fcdonders/fieldtrip/doku.php? id=fieldtrip:documentation:frequently_asked_questions#how_are_the_differ ent_head_and_mri_coordinate_systems_defined for further details on coordinate systems, and the transformation matrix. best regards Robert ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Wed Jan 16 21:13:20 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Wed, 16 Jan 2008 15:13:20 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <265F73253EB9234F81FF4EC2DCB2AD1CA8B25B@nihcesmlbx2.nih.gov> Message-ID: > > ----- Original Message ----- > > From: Peter Lin > > Date: Tuesday, January 8, 2008 9:34 pm > > Subject: [FIELDTRIP] DICS pcc error > > > > > > ??? Error using ==> ctf2grad at 68 > > > unexpected number of coils in MEG channel OK, the ultimate cause of the problem is line 143 in ctf_read_res4.m, which reads: temp=fread(fid,32,'char')'; it should be changed to: temp=fread(fid,32,'uchar')'; and that'll fix it. It is interesting that this works in Octave but not in real Matlab. What happens in Matlab with 'char' is that temp comes back as an array of _shorts_. I have no idea at all why this happens, it's completely wrong behavior. I'd appreciate it if somebody could check me on this. I'm using Matlab 7.2.0.294. With 'uchar' it works as it should. What happens with 'char', in more detail, is that it reads shorts, and the file pointer gets thrown off, and the gains, which are read in next, get read as zero. Also the coil numbers. Dr. Tom -- To them sounds it pleasant to have preached in their ears: "Nothing is worth while! You shall not will!" That, however, is a sermon for slavery. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From hubert.preissl at UNI-TUEBINGEN.DE Thu Jan 17 16:08:26 2008 From: hubert.preissl at UNI-TUEBINGEN.DE (Hubert Preissl) Date: Thu, 17 Jan 2008 16:08:26 +0100 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <1200514400.5186.74.camel@kurage.nimh.nih.gov> Message-ID: hi tom, the error is related to a change in char definition: http://www.mathworks.com/support/solutions/data/1-2782ZU.html?solution=1-2782ZU hubert -- Dr. Hubert Preissl MEG-Center phone: ++49-(0)7071-2987704 Otfried Mueller Str 47 fax: ++49-(0)7071-295706 72076 Tübingen, Germany email: hubert.preissl at uni-tuebingen.de ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From h.morgan at BANGOR.AC.UK Thu Jan 17 18:21:08 2008 From: h.morgan at BANGOR.AC.UK (Helen Morgan) Date: Thu, 17 Jan 2008 18:21:08 +0100 Subject: Post-doc position at Bangor University, UK Message-ID: Research Officer Wolfson Centre For Clinical And Cognitive Neuroscience School of Psychology Starting Salary: £21,478 - £23,883 (on R&A Grade 1A) p.a. Applications are invited for a three-year Postdoctoral Research Officer position in social cognitive neuroscience. This position is to work on a research grant awarded to Professor David Linden jointly with Professor Thomas Mussweiler of the University of Cologne, Germany. This grant is funded by the ESRC through their bilateral programme with the German Research Council (DFG). The aim of the project is to investigate the neural substrates of social comparison, one of the most pervasive cognitive processes humans engage in. The work will involve the adaptation of paradigms from social psychology research for functional imaging and the acquisition and analysis of behavioural and imaging data. The successful candidate will work closely with the research officer in the Cologne team, who is being recruited in parallel. Professor Mussweiler's group in Cologne is internationally leading in the fields of social judgement, stereotyping and decision making: http://social-cognition.uni-koeln.de/. This is a full-time position, starting on April 1st 2008. The ideal candidate would have a PhD in cognitive neuroscience or a related field, experience with stimulus programming and functional magnetic resonance imaging and an interest in the application of these skills to current topics in social psychology. The School of Psychology at Bangor is one of the UK's leading Psychology departments. It enjoys the highest possible research rating (RAE 5*A), and offers an exciting research environment within a dynamic and growing imaging group. Further details of the School, including our extensive research facilities (3T MRI, ERP, MRI-guided TMS), can be found under http://www.psychology.bangor.ac.uk/. The University is situated on the northern coastline of Wales in an area of outstanding natural beauty within a few miles of the Snowdonia Mountains National Park and several quiet beaches. Bangor has direct rail connections to London (3 hours) and Manchester (2 hours). Application forms and further particulars should be obtained by contacting Human Resources, Bangor University; tel: (01248) 382926/388132; e-mail: personnel at bangor.ac.uk; web: www.bangor.ac.uk For an informal discussion about this post, potential applicants are invited to contact Professor David Linden, tel: (01248) 382564, e-mail: d.linden at bangor.ac.uk Please quote reference number 08-7/93 when applying. Closing date for applications: Friday 15th February, 2008. Committed To Equal Opportunities ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 17 20:28:23 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 17 Jan 2008 14:28:23 -0500 Subject: FIELDTRIP Digest - 8 Jan 2008 to 9 Jan 2008 (#2008-5) In-Reply-To: <478F6F6A.7020007@uni-tuebingen.de> Message-ID: Thank you! That was very helpful. I posted on the Octave list that I thought it was great that Octave 3.0 had gotten it right while Matlab had gotten it wrong. Of course now Matlab is saying that a char is 16 bits, breaking old code. Normal people introduce a wchar type to handle 16 bit characters. I hope the Octave people do something sensible, like provide "broken matlab mode" but leave 'char' as 8 bits by default. On Thu, 2008-01-17 at 16:08 +0100, Hubert Preissl wrote: > hi tom, > the error is related to a change in char definition: > > http://www.mathworks.com/support/solutions/data/1-2782ZU.html?solution=1-2782ZU > > > hubert > Dr. Tom -- He has subdued monsters, he has solved enigmas. But he should also redeem his monsters and enigmas; into heavenly children should he transform them. Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tomh at KURAGE.NIMH.NIH.GOV Thu Jan 17 21:00:50 2008 From: tomh at KURAGE.NIMH.NIH.GOV (Tom Holroyd) Date: Thu, 17 Jan 2008 15:00:50 -0500 Subject: char vs uchar Message-ID: The Octave guru suggests that it's better to use 'uint8' as that is unlikely to change. Dr. Tom -- Better, however, to be foolish with happiness than foolish with misfortune, better to dance awkwardly than walk lamely. So learn, I pray you, my wisdom, you higher men: even the worst thing has two good reverse sides, even the worst thing has good dancing-legs: so learn, I pray you, you higher men, to put yourselves on your proper legs! Thus spoke Zarathustra. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From l.frei at PSY.GLA.AC.UK Fri Jan 18 16:24:52 2008 From: l.frei at PSY.GLA.AC.UK (Luisa Frei) Date: Fri, 18 Jan 2008 15:24:52 +0000 Subject: missing files, reshape In-Reply-To: <003001c8576b$a3bfc3b0$642dae83@fcdonders.nl> Message-ID: Ingrid: removing the dof-field did he trick! Jan-Mathijs: thanks for the file! Many thanks to both of you for your help! Luisa On 15 Jan 2008, at 11:41, Ingrid Nieuwenhuis wrote: > Dear Louisa, > > The second problem might also be related to the dof field. If > there's a dof > field in the data, removing it might be a fix. > > Best Ingrid > > -----Original Message----- > From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] > On Behalf > Of jan-mathijs schoffelen > Sent: Tuesday, January 15, 2008 12:20 PM > To: FIELDTRIP at NIC.SURFNET.NL > Subject: Re: [FIELDTRIP] missing files, reshape > > Dear Luisa, > > I put the missing GA_ERF_orig mat-file on the ftp-server. You can > access it at: > > ftp://ftp.fcdonders.nl (username: anonymous, password: your e-mail > address). It's in the directory /pub/temporary. > > As far as I can see, the second problem could be related to the fact > that you do not have the correct data. It might be solved when you > use the data in the mat-file. > > Yours, > > Jan-Mathijs > > > > On Jan 11, 2008, at 1:47 PM, Luisa Frei wrote: > >> Hi, >> I'm currently doing the fieldtrip tutorials and I have encountered >> 2 problems: >> >> 1. Some of the pre-calculated data files are missing. That's >> especially complicating for the Parametric and non-parametric >> statistics tutorial, because the whole tutorial relies on that >> file: GA_ERF_orig.mat >> I also don't have the files for the cluster-based permutation test >> tutorial. Does anyone know, where I can find them? >> >> 2. In the cluster-based permutation test tutorial I get an error >> with the reshape function at the end of 2.1.2.1. Here's the error >> text: >> >>>> [stat] = timelockstatistics(cfg, FICdata, FCdata); >> ??? Error using ==> reshape >> To RESHAPE the number of elements must not change. >> >> Error in ==> fieldtrip-20080107/private/ >> prepare_timefreq_data>forcedimord at 674 >> output.dof=reshape(input.dof, Nrepl, Nfreq, Ntime); >> >> Error in ==> fieldtrip-20080107/private/prepare_timefreq_data at 172 >> [remember{c}, hascrsspctrm] = forcedimord(varargin{c}); >> >> Error in ==> fieldtrip-20080107/private/statistics_wrapper at 199 >> [cfg, data] = prepare_timefreq_data(cfg, varargin{:}); >> >> Error in ==> timelockstatistics at 107 >> [stat] = statistics_wrapper(cfg, varargin{:}); >> >> >> I didn't manage to fix this. Has anyone else encountered this and >> knows how to solve it? Any help would be greatly appreciated. >> >> >> Thanks, >> Luisa >> ---------------------------------- >> The aim of this list is to facilitate the discussion between users >> of the FieldTrip toolbox, to share experiences and to discuss new >> ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ >> archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. >> > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the > FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG > and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. > > ---------------------------------- > The aim of this list is to facilitate the discussion between users > of the FieldTrip toolbox, to share experiences and to discuss new > ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/ > archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From gilbertson.thomas at GOOGLEMAIL.COM Mon Jan 21 19:12:05 2008 From: gilbertson.thomas at GOOGLEMAIL.COM (thomas gilbertson) Date: Mon, 21 Jan 2008 18:12:05 +0000 Subject: topoplot error In-Reply-To: <4B8237E1-3717-431E-A772-D0D94DDDBC03@fcdonders.ru.nl> Message-ID: hi robert, a belated reply. appologies in advance if i have made some really simple mistake here but ive tryed what you suggested and am struggling to implement it successfully. heres what ive done so far:- % here is the variable to multiply mri.tranform with (from volumenormalise) initial = [ 0.0000 -1.0000 0.0000 0.0000 0.9987 0.0000 -0.0517 -32.0000 0.0517 0.0000 0.9987 -54.0000 0.0000 0.0000 0.0000 1.0000 ]; % mri is the imported mri mri.transform = 2 0 0 -92 0 2 0 -128 0 0 2 -74 0 0 0 1 % multiply with the inverse of initial and include in the interpolated source array interp.transform = mri.transform.*initial'; % run volume normalise cfg=[]; cfg.coordinates = 'ctf'; interpN = volumenormalise(cfg, interp); % and i get the following error; ??? Error using ==> colon Maximum variable size allowed by the program is exceeded. Error in ==> spm_smoothto8bit>smoothto8bit at 36 x = -r{i}.s:r{i}.s; ive truncated the error for brevity. thanks for your patience with this. best wishes, tom. ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- An HTML attachment was scrubbed... URL: From linpe at NINDS.NIH.GOV Thu Jan 24 03:14:26 2008 From: linpe at NINDS.NIH.GOV (Peter Lin) Date: Thu, 24 Jan 2008 03:14:26 +0100 Subject: DICS error Message-ID: Recently, there was a problem with running fieldtrip analysis off of the NIMH MEG data for DICS. This worked after a change of ctf_read_res4.m file. I ran the DICS script and it was fine until this error came up: the input is freq data with 273 channels, 21 frequencybins and no timebins reading headmodel from file /home/linpe/raid/linpe/20071031/**_pright.ds/default.hdm using gradiometers specified in the data 3794 dipoles inside, 11074 dipoles outside brain computing cross-spectrum from fourier ??? Reference to non-existent field 'refchan'. Error in ==> prepare_freq_matrices at 247 refindx = match_str(freq.label, cfg.refchan); Error in ==> sourceanalysis at 714 [Cf, Cr, Pr, Ntrials, cfg] = prepare_freq_matrices(cfg, data); Error in ==> pcc1 at 61 [source] = sourceanalysis(cfg, freq); Any idea what led to this? Thanks in advance, Peter ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tillmann at MPIH-FRANKFURT.MPG.DE Fri Jan 25 09:53:31 2008 From: tillmann at MPIH-FRANKFURT.MPG.DE (Christine Tillmann) Date: Fri, 25 Jan 2008 09:53:31 +0100 Subject: Sourcestatistics & MNI-grids Message-ID: Hi All, I'm trying to compute virtual channel data using the lcmv-beamformer. As a first step, I created MNI-aligned dipole grids for all subjects to be able to combine the output of source analysis across subjects. But now it seems a bit tricky to feed those source-level data into sourcestatistics: Sourcestatistics seems to require computing the grand average over subjects first, but using sourcegrandaverage here returns the error message that grid locations are different across subjects. Do I need to interpolate and normalize each volume after all, before combining the source-level data? Maybe by using the transformation matrix that was computed for each subject when I normalised each individual anatomy to the MNI-template?? Or else, by copying the template grid positions to the individual subject data directly?? Can anyone help me out? I'd be grateful for any advice on this! Thanks & Best Regards, Christine ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From tillmann at MPIH-FRANKFURT.MPG.DE Fri Jan 25 16:25:41 2008 From: tillmann at MPIH-FRANKFURT.MPG.DE (Christine Tillmann) Date: Fri, 25 Jan 2008 16:25:41 +0100 Subject: virtual channels Message-ID: Hi everyone, I got rid of the problem I described just before (combining source-level data in sourcestatistics with different grid positions across subjects) by copying the template grid structure to the output of sourceanalysis for each subject, before computing sourcegrandaverage and then doing sourcestatistics. Now, since I want to compute virtual channel data, I'm looking for a way to match the location coordinates of the activated voxels to the filter weights. Does anybody know how to match the indices of specific grid points given in 'inside' and 'filter' to the voxel coordinates given by sourceplot? Best regards, Christine ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From grion at SISSA.IT Mon Jan 28 13:34:14 2008 From: grion at SISSA.IT (Natalia Grion) Date: Mon, 28 Jan 2008 13:34:14 +0100 Subject: Importing Data from Neuralynx Message-ID: Hi All, I want to start with Fieldtrip analysis on animal data recorded with Neuralynx. I have LFP (*.ncs files), spikes (*.nts), and event files (*.nev). So, I have the three wrapper functions; which call these read_xxx functions that detect the file format and subsequently call the appropriate low-level function for each file format. In a first attempt to import data I ran: hdr = read_fcdc_header(file) with a *.ncs file but I found this error: ??? Undefined command/function 'read_neuralynx_ncs'. Error in ==> fieldtrip-20070820\private\read_header at 472 ncs = read_neuralynx_ncs(filename, 1, 0); Error in ==> read_fcdc_header at 37 [hdr] = read_header(varargin{:}); Do you have any explanation for this error? Another question: Are *.nst (spike files) recognized? Thank you all!! ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. From ewingerd at SCIENCE.UVA.NL Mon Jan 28 14:17:55 2008 From: ewingerd at SCIENCE.UVA.NL (Marijn van Wingerden) Date: Mon, 28 Jan 2008 14:17:55 +0100 Subject: Importing Data from Neuralynx In-Reply-To: <479DCBC6.6090203@sissa.it> Message-ID: Hello Natalia, I've been working on getting Neuralynx data into Fieldtrip as well. There are some small and some large problems involved, especially concerning the registration of events. Feel free to e-mail me off this list with specific problems. It seems this particular error occured because you don't have the proper loading functions that the wrapper function calls in your path. They are located in the fileio subfolder, but I'll attach the relevant Nlx functions here in a zip-file. Just put them in your matlab path and the function should be able to run. good luck! marijn van wingerden Natalia Grion wrote: > Hi All, I want to start with Fieldtrip analysis on animal data > recorded with Neuralynx. I have LFP (*.ncs files), spikes (*.nts), and > event files (*.nev). So, I have the three wrapper functions; which > call these read_xxx functions that detect the file format and > subsequently call the appropriate low-level function for each file > format. > In a first attempt to import data I ran: > hdr = read_fcdc_header(file) with a *.ncs file > but I found this error: > ??? Undefined command/function 'read_neuralynx_ncs'. > Error in ==> fieldtrip-20070820\private\read_header at 472 > ncs = read_neuralynx_ncs(filename, 1, 0); > Error in ==> read_fcdc_header at 37 > [hdr] = read_header(varargin{:}); > > Do you have any explanation for this error? > Another question: Are *.nst (spike files) recognized? > > Thank you all!! > > ---------------------------------- > The aim of this list is to facilitate the discussion between users of > the FieldTrip toolbox, to share experiences and to discuss new ideas > for MEG and EEG analysis. See also > http://listserv.surfnet.nl/archives/fieldtrip.html and > http://www.ru.nl/fcdonders/fieldtrip. -- Mr. Marijn van Wingerden, M.Sc. PhD-candidate Animal Physiology & Cognitive Neuroscience (APCN) Centre for NeuroScience (CNS) section of Swammerdam Institute for Life Sciences (SILS) University of Amsterdam, Faculty of Science Kruislaan 320, 1098 SM Amsterdam, The Netherlands Tel: +31 20 525 7713/7547 Fax: +31 20 525 7709 ---------------------------------- The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip. -------------- next part -------------- A non-text attachment was scrubbed... Name: read_neuralynx.zip Type: application/x-zip-compressed Size: 20484 bytes Desc: not available URL: