Input - Look into how `nistats` handles automated BIDS ds ingestion & do that - Is `pyBIDS` still a thing? Output - route all output to `derivatives/IDConn` - can I have subdirectories in `IDConn/`? - should derived stats and wrangled data go in a separate output folder, too?