Hi Reza, 1) We’ve already generated Cohen’s d-style effect size maps for all contrasts, using all subjects, as part of the “Group Average Dataset” available at https://db.humanconnectome.org/data/projects/HCP_1200. If you need it computed for a specific subset of subjects, then yes, you can use the approach that you outlined. Note that the ensuing “effect size” does not account for the family structure in the data (i.e., to the extent that the estimate of the std across subjects is biased by the family structure, then the estimate of the effect size is biased as well).
2) A .dtseries.nii file is still a “spatial map”. We just didn’t bother to formally convert those particular outputs to a .dscalar.nii (e.g., via -cifti-change-mapping). A dscalar version of all the copes (merged into a single file) for a given task and subject are available in the root level of the .feat directory containing the Level2 task analysis results for that task and subject. In newer pipeline versions, we create separate merged files for both the “zstat” and “cope” files of the individual contrasts. However, at the time of the processing of the HCP-YA data, only a single merged dscalar was created, and that was for the copes (and it does not unfortunately have “cope” as part of its filename). 3) We recommend using PALM for group statistical analysis. You can find a tutorial in the “tfMRI and PALM” practical available as part of the HCP Course: https://store.humanconnectome.org/courses/2018/exploring-the-human-connectome.php. And no, you generally do *not* want to use the individual subject “zstat1” maps as inputs to a statistical computation, which would be “computing statistics of a statistic” (rather than the statistic of an effect size). 4) The outputs produced are simply the same as those produced by FSL’s FLAMEO, albeit in CIFTI rather than NIFTI format. So see FSL’s FLAMEO documentation. Cheers, -MH -- Michael Harms, Ph.D. ----------------------------------------------------------- Associate Professor of Psychiatry Washington University School of Medicine Department of Psychiatry, Box 8134 660 South Euclid Ave. Tel: 314-747-6173 St. Louis, MO 63110 Email: mha...@wustl.edu From: <hcp-users-boun...@humanconnectome.org> on behalf of Reza Rajimehr <rajim...@gmail.com> Date: Wednesday, May 29, 2019 at 7:36 PM To: hcp-users <hcp-users@humanconnectome.org> Subject: [HCP-Users] Group-averaging of task fMRI data Hi, For a group of subjects (e.g. 100 subjects in HCP S1200), we want to generate a group-average Cohen’s d map for a particular contrast in the working memory task. For this, we take level2 “cope1.dtseries.nii” file in cope20.feat folder of all those subjects, merge them using -cifti-merge, then -cifti-reduce mean, -cifti-reduce stdev, and -cifti-math mean/stdev. Questions: 1) Is the above procedure correct? Or you recommend other commands? 2) Why the file name is cope1.dtseries.nii when it is not a time-series data? Why not naming it cope1.dscalar.nii, as it is a spatial map? 3) If we want to generate a group-average zstat map, what should we do? I guess it should involve using the “zstat1.dtseries.nii” files, but don’t know how. 4) Is there any documentation somewhere describing all the files within the cope folder? E.g. these files: mean_random_effects_var1.dtseries.nii pe1.dtseries.nii res4d.dtseries.nii tdof_t1.dtseries.nii tstat1.dtseries.nii varcope1.dtseries.nii weights1.dtseries.nii Thanks, Reza _______________________________________________ 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
