[HCP-Users] cifti-resample of 32K dconn to 4K dconn
Hi everyone, i am trying for the MEG practicals to downsample a 32K dconn fMRI group functional connectivity cifti: HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii into a dconn cifti in the 4K space that we use in the MEG analysis in OT4. I am trying to use cifti-resample to do so. I know i have to use it twice, once along COLUMN and then along ROW directions. However I am not sure how to set the cifti-resample input arguments correctly. I have the following files : Input functional connectivity file from fMRI in 32K space: HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii Input Sphere Left (32K space): S900.L.sphere.32k_fs_LR.surf.gii Input Sphere Right (32K space): S900.R.sphere.32k_fs_LR.surf.gii Output Sphere Left (4K space): Sphere.4k.L.surf.gii Output Sphere Right (4K space): Sphere.4k.R.surf.gii I have a "template" (not sure if it is actually suitable for a template) dconn file in the 4K space, called testtemplate4K.dconn.nii. This is a just a dconn file of functional connectivity computed directly from MEG data in the 4K space. The issue is that the 32K functional connectivity matrix contains connectivity between all vertices , in both left and right hemispheres and i am not sure if such a connectivity file can be divided into left and right. So this file is defined over both hemispheres. Then I have the input and output spaces defined with sphere files, different for left and right hemispheres. After reading the description of cifti-resample I tried running the following (with wb_command 1.2): wb_command -cifti-resample HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii COLUMN testtemplate4K.dconn.nii COLUMN ADAP_BARY_AREA CUBIC testdconnresamp1.dconn.gii -left-spheres S900.L.sphere.32k_fs_LR.surf.gii Sphere.4k.L.surf.gii -right-spheres S900.R.sphere.32k_fs_LR.surf.gii Sphere.4k.R.surf.gii and i got the error: ERROR: left area data is missing I thought that the left area data is in the input dconn 32K dconn cifti. Could you please have a look in the above command and advice on what i am doing wrong? Thank you very much Best Giorgos --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ___ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
Re: [HCP-Users] cifti-resample of 32K dconn to 4K dconn
Hi Georgios, I would not recommend downsampling a dense connectome. Instead it would be better to downsample the dense timeseries and then recompute the dense connectome. Peace, Matt. On 7/16/16, 9:47 AM, "hcp-users-boun...@humanconnectome.org on behalf of Georgios Michalareas" wrote: >Hi everyone, > >i am trying for the MEG practicals to downsample a 32K dconn fMRI group >functional connectivity cifti: > >HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii > >into a dconn cifti in the 4K space that we use in the MEG analysis in OT4. > >I am trying to use cifti-resample to do so. I know i have to use it >twice, once along COLUMN and then along ROW directions. > >However I am not sure how to set the cifti-resample input arguments >correctly. > > >I have the following files : > >Input functional connectivity file from fMRI in 32K space: >HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii > >Input Sphere Left (32K space): S900.L.sphere.32k_fs_LR.surf.gii > >Input Sphere Right (32K space): S900.R.sphere.32k_fs_LR.surf.gii > >Output Sphere Left (4K space): Sphere.4k.L.surf.gii > >Output Sphere Right (4K space): Sphere.4k.R.surf.gii > >I have a "template" (not sure if it is actually suitable for a template) >dconn file in the 4K space, called testtemplate4K.dconn.nii. This is a >just a dconn file of functional connectivity computed directly from MEG >data in the 4K space. > >The issue is that the 32K functional connectivity matrix contains >connectivity between all vertices , in both left and right hemispheres >and i am not sure if such a connectivity file can be divided into left >and right. So this file is defined over both hemispheres. > >Then I have the input and output spaces defined with sphere files, >different for left and right hemispheres. > >After reading the description of cifti-resample I tried running the >following (with wb_command 1.2): > >wb_command -cifti-resample >HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii COLUMN >testtemplate4K.dconn.nii COLUMN ADAP_BARY_AREA CUBIC >testdconnresamp1.dconn.gii -left-spheres >S900.L.sphere.32k_fs_LR.surf.gii Sphere.4k.L.surf.gii -right-spheres >S900.R.sphere.32k_fs_LR.surf.gii Sphere.4k.R.surf.gii > >and i got the error: > >ERROR: left area data is missing > >I thought that the left area data is in the input dconn 32K dconn cifti. > >Could you please have a look in the above command and advice on what i >am doing wrong? > >Thank you very much > >Best > >Giorgos > > > > >--- >This email has been checked for viruses by Avast antivirus software. >https://www.avast.com/antivirus > >___ >HCP-Users mailing list >HCP-Users@humanconnectome.org >http://lists.humanconnectome.org/mailman/listinfo/hcp-users The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail. ___ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
Re: [HCP-Users] cifti-resample of 32K dconn to 4K dconn
Thanks for the tip Matt. I ll stick to parcellations for comparing fMRI and MEG connectivity in the practicals. But in any case , if one would want to do such a deconn resampling , do you know what is the mistake(s) I am doing in the command below? Thanks again Giorgos On 7/16/2016 4:49 PM, Glasser, Matthew wrote: > Hi Georgios, > > I would not recommend downsampling a dense connectome. Instead it would > be better to downsample the dense timeseries and then recompute the dense > connectome. > > Peace, > > Matt. > > On 7/16/16, 9:47 AM, "hcp-users-boun...@humanconnectome.org on behalf of > Georgios Michalareas" giorgos.michalar...@esi-frankfurt.de> wrote: > >> Hi everyone, >> >> i am trying for the MEG practicals to downsample a 32K dconn fMRI group >> functional connectivity cifti: >> >> HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii >> >> into a dconn cifti in the 4K space that we use in the MEG analysis in OT4. >> >> I am trying to use cifti-resample to do so. I know i have to use it >> twice, once along COLUMN and then along ROW directions. >> >> However I am not sure how to set the cifti-resample input arguments >> correctly. >> >> >> I have the following files : >> >> Input functional connectivity file from fMRI in 32K space: >> HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii >> >> Input Sphere Left (32K space): S900.L.sphere.32k_fs_LR.surf.gii >> >> Input Sphere Right (32K space): S900.R.sphere.32k_fs_LR.surf.gii >> >> Output Sphere Left (4K space): Sphere.4k.L.surf.gii >> >> Output Sphere Right (4K space): Sphere.4k.R.surf.gii >> >> I have a "template" (not sure if it is actually suitable for a template) >> dconn file in the 4K space, called testtemplate4K.dconn.nii. This is a >> just a dconn file of functional connectivity computed directly from MEG >> data in the 4K space. >> >> The issue is that the 32K functional connectivity matrix contains >> connectivity between all vertices , in both left and right hemispheres >> and i am not sure if such a connectivity file can be divided into left >> and right. So this file is defined over both hemispheres. >> >> Then I have the input and output spaces defined with sphere files, >> different for left and right hemispheres. >> >> After reading the description of cifti-resample I tried running the >> following (with wb_command 1.2): >> >> wb_command -cifti-resample >> HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii COLUMN >> testtemplate4K.dconn.nii COLUMN ADAP_BARY_AREA CUBIC >> testdconnresamp1.dconn.gii -left-spheres >> S900.L.sphere.32k_fs_LR.surf.gii Sphere.4k.L.surf.gii -right-spheres >> S900.R.sphere.32k_fs_LR.surf.gii Sphere.4k.R.surf.gii >> >> and i got the error: >> >> ERROR: left area data is missing >> >> I thought that the left area data is in the input dconn 32K dconn cifti. >> >> Could you please have a look in the above command and advice on what i >> am doing wrong? >> >> Thank you very much >> >> Best >> >> Giorgos >> >> >> >> >> --- >> This email has been checked for viruses by Avast antivirus software. >> https://www.avast.com/antivirus >> >> ___ >> HCP-Users mailing list >> HCP-Users@humanconnectome.org >> http://lists.humanconnectome.org/mailman/listinfo/hcp-users > > > The materials in this message are private and may contain Protected > Healthcare Information or other information of a sensitive nature. If you are > not the intended recipient, be advised that any unauthorized use, disclosure, > copying or the taking of any action in reliance on the contents of this > information is strictly prohibited. If you have received this email in error, > please immediately notify the sender via telephone or return mail. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ___ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
Re: [HCP-Users] cifti-resample of 32K dconn to 4K dconn
Hi Georgios, Looks like you asked for adaptive barycentric resampling without providing areas to use (the algorithm does not accept spherical surfaces to be the source of anatomical vertex areas). Resampling a dense connectome would only be appropriate for tractography data. For fMRI data, you want to average the timeseries prior to running functional connectivity at the lower resolution. Peace, Matt. On 7/16/16, 9:59 AM, "Georgios Michalareas" wrote: >Thanks for the tip Matt. > >I ll stick to parcellations for comparing fMRI and MEG connectivity in >the practicals. > >But in any case , if one would want to do such a deconn resampling , do >you know what is the mistake(s) I am doing in the command below? > >Thanks again > >Giorgos > > >On 7/16/2016 4:49 PM, Glasser, Matthew wrote: >> Hi Georgios, >> >> I would not recommend downsampling a dense connectome. Instead it would >> be better to downsample the dense timeseries and then recompute the >>dense >> connectome. >> >> Peace, >> >> Matt. >> >> On 7/16/16, 9:47 AM, "hcp-users-boun...@humanconnectome.org on behalf of >> Georgios Michalareas" >of >> giorgos.michalar...@esi-frankfurt.de> wrote: >> >>> Hi everyone, >>> >>> i am trying for the MEG practicals to downsample a 32K dconn fMRI group >>> functional connectivity cifti: >>> >>> HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii >>> >>> into a dconn cifti in the 4K space that we use in the MEG analysis in >>>OT4. >>> >>> I am trying to use cifti-resample to do so. I know i have to use it >>> twice, once along COLUMN and then along ROW directions. >>> >>> However I am not sure how to set the cifti-resample input arguments >>> correctly. >>> >>> >>> I have the following files : >>> >>> Input functional connectivity file from fMRI in 32K space: >>> HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii >>> >>> Input Sphere Left (32K space): S900.L.sphere.32k_fs_LR.surf.gii >>> >>> Input Sphere Right (32K space): S900.R.sphere.32k_fs_LR.surf.gii >>> >>> Output Sphere Left (4K space): Sphere.4k.L.surf.gii >>> >>> Output Sphere Right (4K space): Sphere.4k.R.surf.gii >>> >>> I have a "template" (not sure if it is actually suitable for a >>>template) >>> dconn file in the 4K space, called testtemplate4K.dconn.nii. This is a >>> just a dconn file of functional connectivity computed directly from MEG >>> data in the 4K space. >>> >>> The issue is that the 32K functional connectivity matrix contains >>> connectivity between all vertices , in both left and right hemispheres >>> and i am not sure if such a connectivity file can be divided into left >>> and right. So this file is defined over both hemispheres. >>> >>> Then I have the input and output spaces defined with sphere files, >>> different for left and right hemispheres. >>> >>> After reading the description of cifti-resample I tried running the >>> following (with wb_command 1.2): >>> >>> wb_command -cifti-resample >>> HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii COLUMN >>> testtemplate4K.dconn.nii COLUMN ADAP_BARY_AREA CUBIC >>> testdconnresamp1.dconn.gii -left-spheres >>> S900.L.sphere.32k_fs_LR.surf.gii Sphere.4k.L.surf.gii -right-spheres >>> S900.R.sphere.32k_fs_LR.surf.gii Sphere.4k.R.surf.gii >>> >>> and i got the error: >>> >>> ERROR: left area data is missing >>> >>> I thought that the left area data is in the input dconn 32K dconn >>>cifti. >>> >>> Could you please have a look in the above command and advice on what i >>> am doing wrong? >>> >>> Thank you very much >>> >>> Best >>> >>> Giorgos >>> >>> >>> >>> >>> --- >>> This email has been checked for viruses by Avast antivirus software. >>> https://www.avast.com/antivirus >>> >>> ___ >>> HCP-Users mailing list >>> HCP-Users@humanconnectome.org >>> http://lists.humanconnectome.org/mailman/listinfo/hcp-users >> >> >> The materials in this message are private and may contain Protected >>Healthcare Information or other information of a sensitive nature. If >>you are not the intended recipient, be advised that any unauthorized >>use, disclosure, copying or the taking of any action in reliance on the >>contents of this information is strictly prohibited. If you have >>received this email in error, please immediately notify the sender via >>telephone or return mail. > > >--- >This email has been checked for viruses by Avast antivirus software. >https://www.avast.com/antivirus > The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail. ___
Re: [HCP-Users] cifti-resample of 32K dconn to 4K dconn
Thanks On 7/16/2016 5:03 PM, Glasser, Matthew wrote: > Hi Georgios, > > Looks like you asked for adaptive barycentric resampling without providing > areas to use (the algorithm does not accept spherical surfaces to be the > source of anatomical vertex areas). Resampling a dense connectome would > only be appropriate for tractography data. For fMRI data, you want to > average the timeseries prior to running functional connectivity at the > lower resolution. > > Peace, > > Matt. > > On 7/16/16, 9:59 AM, "Georgios Michalareas" > wrote: > >> Thanks for the tip Matt. >> >> I ll stick to parcellations for comparing fMRI and MEG connectivity in >> the practicals. >> >> But in any case , if one would want to do such a deconn resampling , do >> you know what is the mistake(s) I am doing in the command below? >> >> Thanks again >> >> Giorgos >> >> >> On 7/16/2016 4:49 PM, Glasser, Matthew wrote: >>> Hi Georgios, >>> >>> I would not recommend downsampling a dense connectome. Instead it would >>> be better to downsample the dense timeseries and then recompute the >>> dense >>> connectome. >>> >>> Peace, >>> >>> Matt. >>> >>> On 7/16/16, 9:47 AM, "hcp-users-boun...@humanconnectome.org on behalf of >>> Georgios Michalareas" >> of >>> giorgos.michalar...@esi-frankfurt.de> wrote: >>> Hi everyone, i am trying for the MEG practicals to downsample a 32K dconn fMRI group functional connectivity cifti: HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii into a dconn cifti in the 4K space that we use in the MEG analysis in OT4. I am trying to use cifti-resample to do so. I know i have to use it twice, once along COLUMN and then along ROW directions. However I am not sure how to set the cifti-resample input arguments correctly. I have the following files : Input functional connectivity file from fMRI in 32K space: HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii Input Sphere Left (32K space): S900.L.sphere.32k_fs_LR.surf.gii Input Sphere Right (32K space): S900.R.sphere.32k_fs_LR.surf.gii Output Sphere Left (4K space): Sphere.4k.L.surf.gii Output Sphere Right (4K space): Sphere.4k.R.surf.gii I have a "template" (not sure if it is actually suitable for a template) dconn file in the 4K space, called testtemplate4K.dconn.nii. This is a just a dconn file of functional connectivity computed directly from MEG data in the 4K space. The issue is that the 32K functional connectivity matrix contains connectivity between all vertices , in both left and right hemispheres and i am not sure if such a connectivity file can be divided into left and right. So this file is defined over both hemispheres. Then I have the input and output spaces defined with sphere files, different for left and right hemispheres. After reading the description of cifti-resample I tried running the following (with wb_command 1.2): wb_command -cifti-resample HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii COLUMN testtemplate4K.dconn.nii COLUMN ADAP_BARY_AREA CUBIC testdconnresamp1.dconn.gii -left-spheres S900.L.sphere.32k_fs_LR.surf.gii Sphere.4k.L.surf.gii -right-spheres S900.R.sphere.32k_fs_LR.surf.gii Sphere.4k.R.surf.gii and i got the error: ERROR: left area data is missing I thought that the left area data is in the input dconn 32K dconn cifti. Could you please have a look in the above command and advice on what i am doing wrong? Thank you very much Best Giorgos --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ___ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users >>> >>> The materials in this message are private and may contain Protected >>> Healthcare Information or other information of a sensitive nature. If >>> you are not the intended recipient, be advised that any unauthorized >>> use, disclosure, copying or the taking of any action in reliance on the >>> contents of this information is strictly prohibited. If you have >>> received this email in error, please immediately notify the sender via >>> telephone or return mail. >> >> --- >> This email has been checked for viruses by Avast antivirus software. >> https://www.avast.com/antivirus >> > > > The materials in this message are private and may contain Protected > Healthcare Information or other information of a sensitive nature. If you are > not the intended recipient, be advised that any unauthorized use, disclosure, > copying or the taking of a
[HCP-Users] computing pairwise correlations between pconn files
Hi again, another question. I am trying to compute pairwise correlation with -cifti-pairwise-correlate between two pconn files. One has been derived by parcellating (-cifti-parcellate) with a given parcellation , i.e. Yeo11, the dconn file in 32K space , i.e. HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii resulting in the pconn file: HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.Yeo11.pconn.nii The other file has resulted from parcellating a dconn file describing group average functional connectivity from MEG data in the 4K space used for MEG analysis,i.e. SUBJAVG_MEG-Restin_icaimagcoh_alpha.Yeo11.pconn.nii So while the two pconn files represent connectivity between the same parcels, they have been derived by parcellating different spaces(32K and 4K ). When I compute the pairwise correlation as: wb_command -cifti-pairwise-correlation HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.Yeo11.pconn.nii SUBJAVG_MEG-Restin_icaimagcoh_alpha.Yeo11.pconn.nii outConneCorrel.nii I get the error: ERROR: mapping along columns must match between the input files Do you know how I could hack this and compute the correlation between the two pconn matrices as if they had been parcellated from the same spatial representation so that it would also be used in the output pscalar file ? Thanks again for your help Best Giorgos These pconn files represent fucntional connectivia simple comparison between fmri mattr --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ___ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
[HCP-Users] HCP script used for generating group-averaged dense connectome
Dear HCP experts, Could I find the shell or MATLAB script used for generating group-averaged dense connectome ("HCP_S900_820_rfMRI_MSMAll_groupPCA_d4500ROW_zcorr.dconn.nii") somewhere online? Thank you. ___ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users