Hi Michael and Greg, Great, thank you for getting back to me so quickly! Your responses and the new Wiki entry are very helpful. Will keep an eye out for the level 3 script when it debuts, but should be able to proceed for now based on what you¹ve told me.
Thanks again, -Ely On 8/28/14, 13:44, "Greg Burgess" <[email protected]> wrote: >Hello Ben, > >Since this process has been causing some confusion among multiple users, >we¹ve decided to put some additional detail in a wiki page. >https://urldefense.proofpoint.com/v1/url?u=https://wiki.humanconnectome.or >g/display/PublicData/Advice%2Bfor%2BFEAT%2BAnalysis%2Bof%2BHCP%2Btask%2BfM >RI%2Bdata&k=PFsgZv%2B4TEaU9fTpAWpLNw%3D%3D%0A&r=uVl%2BXwg39nHsh67CdvztQqky >iHZM2N%2FMMLz%2FhfhLnXM%3D%0A&m=YbJyXW%2BzAfmphLjU2H5hXd82w9hpk2SwexKyasvY >sZ4%3D%0A&s=946b127d507d32ba29d3128c86e32ca1a5c8b7ba22e7091fea9a8ebd2f0689 >1d > >Please review this page to determine if it clarifies the process that you >should follow to run group-level analyses of task fMRI data. If you have >any additional questions, let us know so that we can further clarify this >for users. > >Thanks, >--Greg > >____________________________________________________________________ >Greg Burgess, Ph.D. >Staff Scientist, Human Connectome Project >Washington University School of Medicine >Department of Anatomy and Neurobiology >Phone: 314-362-7864 >Email: [email protected] > >On Aug 28, 2014, at 9:48 AM, Harms, Michael <[email protected]> wrote: > >> >> Hello Ely, >> If you want to run a "Level 3" group analysis, you already have >>everything you need if you downloaded the task "analysis" packages, >>which already contain the "Level 2" results (combining across runs) for >>each subject. The "prepare_level2_feat_analysis.sh" is if you are >>trying to generate the Level 2 analysis for each subject yourself, from >>"Level 1" results (each specific run). The Level 1 results are not >>included in the downloadable packages. >> >> In terms of running the Level 3 analysis, we have a script that helps >>automate this, but it needs a little more vetting and clean-up, so I'm >>not sure if it will quite make it into the imminent release of the HCP >>pipelines. For a volume-based analysis (i.e., using the results in the >>tfMRI_<TASK>_hp200_s4_level2vol.feat directory for each subject), I >>believe that you can run the Level 3 analysis through the FEAT GUI. >>(Just tested that on an simple three-subject group mean model, and it >>worked). For a grayordinate-based analysis, you basically have to >>convert the CIFTI to NIFTI using 'wb_command -cifti-convert', run the >>equivalent set of commands implemented by FEAT on the ensuing NIFTI's, >>and then convert back to CIFTI. The aforementioned Level3 analysis >>script is obviously intended to make this much simpler, and we will try >>to get that vetted and ready for release in the near future. >> >> cheers, >> -MH >> >> -- >> Michael Harms, Ph.D. >> ----------------------------------------------------------- >> Conte Center for the Neuroscience of Mental Disorders >> Washington University School of Medicine >> Department of Psychiatry, Box 8134 >> 660 South Euclid Ave. Tel: 314-747-6173 >> St. Louis, MO 63110 Email: [email protected] >> >> From: <Ely>, Benjamin <[email protected]> >> Date: Thursday, August 28, 2014 1:20 AM >> To: "[email protected]" <[email protected]> >> Subject: [HCP-Users] missing level 1 .feat directories >> >> Hello all, >> >> I recently downloaded the preprocessed and analyzed volume data from >>the 100 Unrelated Subjects group. My understanding is that to conduct >>group analyses, I now need to run the prepare_level2_feat_analysis.sh >>script for each of the tasks for each of the subjects. However, when I >>try to run that script from the various level 2 task directories for >>various subjects, I always get the same error message, e.g.: >> >> Cannot find >>/Volumes/Connectomizer/HCP/Subjects/111312/MNINonLinear/Results/tfMRI_GAM >>BLING/../tfMRI_GAMBLING_RL/tfMRI_GAMBLING_RL_hp200_s4.feat >> Cannot find >>/Volumes/Connectomizer/HCP/Subjects/111312/MNINonLinear/Results/tfMRI_GAM >>BLING/../tfMRI_GAMBLING_LR/tfMRI_GAMBLING_LR_hp200_s4.feat >> Cannot find any lower-level .feat directories!!! >> >> When I look in those lower-level task directories, I only see the >>level1.fsf files, so it seems like the level 1 .feat directories are >>missing. These level 1 .feat directories are not mentioned in the >>directory structure information from the 500 Subjects Release Reference >>Manual, though. >> >> Are these level 1 .feat directories necessary, and if so, is there a >>way to download them? Or will I need to generate them using FSL? >> >> Many thanks for your help! >> -Ely >> >> >> B. Ely >> PhD Student >> Icahn School of Medicine at Mount Sinai >> _______________________________________________ >> HCP-Users mailing list >> [email protected] >> >>https://urldefense.proofpoint.com/v1/url?u=http://lists.humanconnectome.o >>rg/mailman/listinfo/hcp-users&k=PFsgZv%2B4TEaU9fTpAWpLNw%3D%3D%0A&r=uVl%2 >>BXwg39nHsh67CdvztQqkyiHZM2N%2FMMLz%2FhfhLnXM%3D%0A&m=YbJyXW%2BzAfmphLjU2H >>5hXd82w9hpk2SwexKyasvYsZ4%3D%0A&s=510cc6dc82eedad4024d52bb894ada109aa2dad >>f67462327cec51efe62d97c23 >> >> >> >> >> 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 >> [email protected] >> >>https://urldefense.proofpoint.com/v1/url?u=http://lists.humanconnectome.o >>rg/mailman/listinfo/hcp-users&k=PFsgZv%2B4TEaU9fTpAWpLNw%3D%3D%0A&r=uVl%2 >>BXwg39nHsh67CdvztQqkyiHZM2N%2FMMLz%2FhfhLnXM%3D%0A&m=YbJyXW%2BzAfmphLjU2H >>5hXd82w9hpk2SwexKyasvYsZ4%3D%0A&s=510cc6dc82eedad4024d52bb894ada109aa2dad >>f67462327cec51efe62d97c23 >> > _______________________________________________ HCP-Users mailing list [email protected] http://lists.humanconnectome.org/mailman/listinfo/hcp-users
