Re: [HCP-Users] Probtrackx connectivity matrices

[Glasser, Matthew]

That makes sense, particularly as one gets closer to the cortex.  I do think Oxford has some algorithm that checks the fiber orientations across subjects and swaps their order (to make them maximally matching) for their bedpostx TBSS.  TBSS only cares about aligning the cores of tracts that are further away from cortex (and where one might hope to achieve correspondence).   Matt. From: Timothy Coalson <tsc...@mst.edu> Date: Friday, May 9, 2014 at 5:43 PM To: Matt Glasser <glass...@wusm.wustl.edu> Cc: "Parisot, Sarah" <s.pari...@imperial.ac.uk>, "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] Probtrackx connectivity matrices Well, comparison and averaging are not quite the same thing - the idea of averaging matrix4 presents quite a few challenges, and same voxel space is not the messiest - it is more difficult to reconcile anatomy that isn't (and/or can't be) aligned, so the fiber orientations won't match, or places where only some subjects have a second or third fiber orientation. In other words, don't expect matrix4 to be able to be able to be averaged across subjects. I expect that aligning matrix4 to an atlas space is more useful for just looking at the display of tracts of different subjects at the same time. Tim On Fri, May 9, 2014 at 5:31 PM, Glasser, Matthew <glass...@wusm.wustl.edu> wrote: How would one then average across subjects (if that is why they wanted data in standard space)? Matt. From: Timothy Coalson <tsc...@mst.edu> Date: Friday, May 9, 2014 at 3:12 PM To: Matt Glasser <glass...@wusm.wustl.edu> Cc: "Parisot, Sarah" <s.pari...@imperial.ac.uk>, "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] Probtrackx connectivity matrices I don't think we currently have a command to warp the fiber orientations into a different space.  It is possible in theory, and the matrix4 wouldn't need to be regenerated, but the fiber drawing locations won't line up with voxel centers anymore (this is cosmetic more than anything, but it bothers some people). Tim On Thu, May 8, 2014 at 12:36 PM, Glasser, Matthew <glass...@wusm.wustl.edu> wrote: For matrix1, you already have it on a standard mesh that already is comparable across subjects (assuming it is surface data).  For matrix4, I am wondering this myself, however if Tim Coalson does not know, the other person who might is not available right now.  I¹m actually not sure how matrix4 would work in MNI space, because it is dependent on knowing which bedpostx voxels fiber orientations were used for every whiteordinate. There might be more development we have to do to enable this capability. Peace, Matt. On 5/8/14, 12:32 PM, "Parisot, Sarah" <s.pari...@imperial.ac.uk> wrote: >Dear all, > >Thank you for all the help. >I have obtained the tracts and connectivity matrices (matrices 4 and 1) >in the Native domain (32k surface), and would now like to be able to >transfer it to the MNI space for inter subject comparison. >Is it possible to register the output of probtrackx to the MNI (in >particular the matrix4 that saves the tracts) or do I need to run >probtrackx again and specify the native to MNI transformation? >What is the correspondence between the vertices of the 32k native surface >and the 32k MNI surface ? Is it possible to visualise the connectivity on >the MNI surface or do I need to go through a registration step? > >Thanks, >Sarah > >On 15 Apr 2014, at 13:50, Glasser, Matthew <glass...@wusm.wustl.edu> >wrote: > >> Did you use the surfaces in >>${StudyFolder}/${Subject}/T1w/fsaverage_LR32k >> and is your diffusion data in ${StudyFolder}/${Subject}/T1w/Diffusion? >> >> The metric file should be able to be displayed on the same surfaces you >> use for tractography.  Here is code that I use for the conversions: >> >> if [ $MatrixNumber = "1" ] ; then >>      MakeSymmetric="" >>    elif [ $MatrixNumber = "3" ] ; then >>      MakeSymmetric="-make-symmetric" >>    fi >> >> >>Dir="${ResultsFolder}/${Hemisphere}_Trajectory_Matrix${MatrixNumber}${PDD >>ir >> }_${StepSize}" >> $Caret7_Command -probtrackx-dot-convert >> ${Dir}/fdt_matrix${MatrixNumber}.dot >> ${Dir}/fdt_matrix"$MatrixNumber".dconn.nii -row-surface >> ${Dir}/SeedSpaceMetric.func.gii -col-surface >> ${Dir}/SeedSpaceMetric.func.gii -transpose ${MakeSymmetric} >>    $Caret7_Command -convert-matrix4-to-workbench-sparse >> ${Dir}/fdt_matrix4_1.mtx ${Dir}/fdt_matrix4_2.mtx >>${Dir}/fdt_matrix4_3.mtx >> >>${BedpostXFolder}/Whole_Brain_Trajectory_${DiffusionResolution}.fiberTEMP >>.n >> ii ${Dir}/tract_space_coords_for_fdt_matrix4 -surface-seeds >> ${Dir}/SeedSpaceMetric.func.gii ${Dir}/fdt_matrix4.trajTEMP.wbsparse >>    $Caret7_Command -convert-matrix4-to-matrix2 >> ${Dir}/fdt_matrix4.trajTEMP.wbsparse ${Dir}/fdt_matrix2.dconn.nii >> >> The latest version of workbench has visualization for all of these file >> types, and the next workbench tutorial should cover them. >> >> >> Peace, >> >> Matt. >> >> On 4/15/14, 5:42 AM, "Sarah Parisot" <s.pari...@imperial.ac.uk> wrote: >> >>> Hi Matt, >>> >>> Here is my probtrack call. I ran it for the left hemisphere. >>> >>> probtrackx2 --samples=merged >>> --mask=/T1w/L_Cerebral_Trajectory_1.25.nii.gz >>> --seed=100307.L.white.32k_fs_LR.surf.gii >>> --waypoints=100307.L.white.32k_fs_LR.surf.gii >>> --stop=100307.L.pial.32k_fs_LR.surf.gii >>> --target4=L_Cerebral_Trajectory_1.25.nii.gz >>> --seedref=L_Cerebral_Trajectory_1.25.nii.gz >>> --dir=/data/probtrack100307/left --nsamples=5000 --cthr=0.2 >>> --steplength=0.3 --forcedir --meshspace=caret --nsteps=2000 >>> --fibthresh=0.05 --loopcheck --randfib=2 --forcefirststep --opd >>> --omatrix1 --omatrix4 --sampvox=1.25 --verbose=1 >>> >>> Thanks, >>> Sarah >>> >>> On 14/04/14 19:05, Glasser, Matthew wrote: >>>> Can you send us your probtrackx call? >>>> >>>> Peace, >>>> >>>> Matt. >>>> >>>> On 4/14/14, 1:01 PM, "Sarah Parisot" <s.pari...@imperial.ac.uk> wrote: >>>> >>>>> Dear all, >>>>> >>>>> I am working on brain parcellation based on structural connectivity >>>>> matrices. I ran FSL's probtrackx on an HCP subject, seeding from the >>>>> WM/GM 32k surface to get the seedxseed connectivity matrix (matrix1) >>>>> and >>>>> I am struggling a little with the outputs of probtrackx and the use >>>>>of >>>>> the workbench for visualization. It would be great if I could have >>>>>some >>>>> answers to a few questions: >>>>> >>>>> -I obtained a matrix1.dot file that I would like to be able to >>>>> visualize >>>>> using the connectome workbench. I tried running the >>>>> probtrackx-dot-convert command (I used a metric file with all ones >>>>>for >>>>> all nodes as -row/col-surface.) but the converted file appears to be >>>>> all >>>>> zeros when I load it in the workbench. Maybe I'm using the wrong >>>>>metric >>>>> file? >>>>> >>>>> -Is there a way to keep knowledge of the different tracts evaluated >>>>> from >>>>> one seed point to another? >>>>> I am getting an image that counts all the tracts that go through a >>>>> voxel, but I would like to know how seed A is connected to seed B. I >>>>> was >>>>> told this information is kept in the 'matrix4' output, which >>>>>apparently >>>>> cannot be read at the moment. >>>>> >>>>> Thanks. >>>>> Best Regards, >>>>> Sarah >>>>> _______________________________________________ >>>>> 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. >>>> >>> >> >> >> ________________________________ >> 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. > ________________________________ 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   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.   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