Hi,

I want to know if cc200.1D in abide can be processed from the raw data of NII, and what the process is. I tried using functions such as NiftiBabelsMasker in Nilearn to process cpac/filt_global/func_reproc/Saltech_0051458_func_preproc.ni.gz, but the time series results obtained were inconsistent with the downloaded 1d file data. If you could answer, I would be very grateful.

Hi Cherry,

1. resample the preprocessed file into the space of the atlas (flirt -in Caltech_0051458_func_preproc.nii.gz -ref cc200_roi_atlas.nii.gz -out Caltech_0051458_func_preproc_flirt.nii.gz -omat Caltech_0051458_func_preproc_flirt.mat -applyxfm -init ident.mat -interp trilinear)
   
2. run this Python function to extract mean of voxel across all timepoints for each node in the ROI mask. That code is pretty old and in Python 2, but by bringing it up to Python 3 as shown in the attached file:
   

The 1D file generated from that function (with the output from step 1 as data_file and cc200_roi_atlas.nii.gz from s3://fcp-indi/data/Projects/ABIDE/Resources/cc200_roi_atlas.nii.gz as template) is identical to s3://fcp-indi/data/Projects/ABIDE/Outputs/cpac/filt_global/rois_cc200/Caltech_0051458_rois_cc200.1D

Our team can continue looking into other reasons why a current C-PAC image wouldn't generate the same 1D file - let me know if this helps in the meantime!