niflow-nbclab-idconn

Specfify input files

Eventually, this pipeline will read in a BIDS-derivatives directory. For the time being, you will be asked to specify a preprocessed fMRI nifi and, if running an analysis in native space, a T1w nifti per run. In order to accommodate a variety of file organizations you will be expected to specify a generic file path template in which subject, run, session, and task are specifed. These values will be overwritten by the subject IDs, run numbers, sessions, and tasks specified elsewhere.
You will also be asked to specify an atlas or parcellation to use for nodes in the event of a graph-based analysis via --parcellation=. Not necessary in the event of a seed-to-voxel analysis, in which case you will use the flag --roi= to point to the seed you intend to use for analysis.

Specify algorithm to use to calculate edge weights

Accepts any of the methods included in scipy.stats.pairwise or "pearsonr" to use the Pearson product moment correlation to calculate edge weights via scipy.stats.pearsonr. Default is "pearsonr".

Option to compute graph theoretic metrics from connectivity graphs.

If you'd like to use graph theoretic metrics to compute summary statistics from connectivity graphs, please specify using --graph=True. Specify metrics using --metrics tag, using any of the metrics specified in bctpy.

Output will be written in .csv files in the specified directory

Using --output=, specify the output directory in which metrics will be written out by subject in .csv files. If multiple runs and/or sessions are specified, output will be MultiIndexed(e.g., subjectXrunXsession).