Group ICA - Maximum Number of IC Allowed as 2 #27
Description
When trying to run a group ICA using GIFT, for some reason it says we cannot use more than 2 components in our ICA despite the fact that the component estimate from GIFT is saying we should have 78 components. Oddly enough we have a seperate dataset collected from a different study with relatively the same MRI parameter's (except with a 2mm instead of 3mm voxel) that we had done a group ICA through GIFT on previously and can still run the same group analysis in our computing environment. We did have to update the folder structure of the old dataset to be study/sub-XXX/func/BOLD_file.nii per file from study/sub-XXX/BOLD_file.nii to get it to work though.
I've tried a handful of solutions like switching between BIDS and non-BIDS (the structure that works for our older study) folder structure, comparing our data to the old dataset that it works on, and fixing seemingly irrelavent but incorrect BOLD metadata in the header (primarily the 'extends' folder originally being set to 0 instead of 16384) but nothing seems to work. The one oddity that we can't seem to resolve is that within our BOLD metadata the 'vox_offset' seems to only be settable to 0. If I try and increase it to our actual vox_offset it overwrites it to 0. That being said in our older dataset that we can get group ICA to run the files also have vox_offset set to 0 so I'm hesitant to say this is the issue, but I'm at a loss.
Do you have any suggestions on how to resolve this issue? Thanks for your time and help on this!