[ENH] Output derivatives to folder outside of BIDS dataset

.github/pull_request_template.md

@@ -0,0 +1,23 @@
+<!---
+This is a suggested pull request template for cis-processing.
+It's designed to capture information we've found to be useful in reviewing pull requests.
+
+If there is other information that would be helpful to include, please don't hesitate to add it!
+
+Please also label your pull request with the relevant tags.
+Here is a list of appropriate tags:
+[ENH]: Adds a new feature
+[FIX]: Fixes a bug
+[REF]: Refactors (reorganizes, improves, etc.) code without impacting behavior
+[STY]: Improves coding style
+[DOC]: Improvements to documentation, including commenting
+[TST]: Testing (e.g., adding unit/integration tests)
+[WIP]: A work in progress. We may discuss this PR but will not merge until it is ready
+-->
+
+Closes # .
+
+Changes proposed in this pull request:
+
+-
+-

cis_proc.py

@@ -13,6 +13,7 @@
 
 import argparse
 import pandas as pd
+from utils import merge_bids
 
 
 def run(command, env={}):
@@ -103,8 +104,9 @@ def main(argv=None):
                          config_options['mriqc'])
     mriqc_version = op.basename(mriqc_file).split('-')[0].split('_')[-1]
 
-    out_deriv_dir = op.join(args.bids_dir,
-                            'derivatives/mriqc-{0}'.format(mriqc_version))
+    # Output derivatives to parent directory of BIDS dataset directory.
+    out_deriv_dir = op.abspath(op.join(
+        args.bids_dir, '../derivatives/mriqc-{0}'.format(mriqc_version)))
 
     # Additional checks and copying for heuristics file
     heuristics_file = config_options['heuristics']
@@ -174,47 +176,8 @@ def main(argv=None):
     os.remove(op.join(scan_work_dir, 'validator.txt'))
 
     if bids_successful:
-        # Merge BIDS dataset into final folder
-        dset_files = ['CHANGES', 'README', 'dataset_description.json',
-                      'participants.tsv']
-        for dset_file in dset_files:
-            if not op.isfile(op.join(args.bids_dir, dset_file)):
-                shutil.copyfile(op.join(scan_work_dir, 'bids', dset_file),
-                                op.join(args.bids_dir, dset_file))
-
-        p_df = pd.read_csv(op.join(scan_work_dir, 'bids/participants.tsv'),
-                           sep='\t')
-        p_df = p_df.T.drop_duplicates().T
-        p_df2 = pd.read_csv(op.join(args.bids_dir, 'participants.tsv'),
-                            sep='\t')
-        p_df2 = p_df2.T.drop_duplicates().T
-
-        # Check if row already in participants file
-        matches = p_df[(p_df == p_df2.loc[0]).all(axis=1)]
-        match = matches.index.values.size
-        if not match:
-            p_df = pd.concat((p_df, p_df2))
-            p_df.to_csv(op.join(args.bids_dir, 'participants.tsv'),
-                        sep='\t', index=False)
-        else:
-            print('Subject/session already found in participants.tsv')
-
-        scratch_sub_dir = op.join(scan_work_dir,
-                                  'bids/sub-{0}'.format(args.sub))
-        out_sub_dir = op.join(args.bids_dir, 'sub-{0}'.format(args.sub))
-        if not op.isdir(out_sub_dir):
-            shutil.copytree(scratch_sub_dir, out_sub_dir)
-        elif args.ses is not None:
-            scratch_ses_dir = op.join(scratch_sub_dir,
-                                      'ses-{0}'.format(args.ses))
-            out_ses_dir = op.join(out_sub_dir, 'ses-{0}'.format(args.ses))
-            if not op.isdir(out_ses_dir):
-                shutil.copytree(scratch_ses_dir, out_ses_dir)
-            else:
-                print('Warning: Subject/session directory already exists in '
-                      'dataset.')
-        else:
-            print('Warning: Subject directory already exists in dataset.')
+        merge_bids(op.join(scan_work_dir, 'bids'), args.bids_dir,
+                   args.sub, args.ses)
 
         # Run MRIQC
         if not op.isdir(out_deriv_dir):

utils.py

@@ -0,0 +1,52 @@
+"""
+Utility functions for cis-processing tools.
+"""
+import shutil
+import os.path as op
+import pandas as pd
+
+
+def merge_bids(in_dir, dset_dir, sub, ses=None):
+    """
+    Merge the single-subject BIDS dataset generated by the cis_proc pipeline
+    into the general project BIDS dataset.
+    """
+    dset_files = ['CHANGES', 'README', 'dataset_description.json',
+                  'participants.tsv']
+    for dset_file in dset_files:
+        if not op.isfile(op.join(dset_dir, dset_file)):
+            shutil.copyfile(op.join(in_dir, dset_file),
+                            op.join(dset_dir, dset_file))
+
+    p_df = pd.read_csv(op.join(in_dir, 'participants.tsv'),
+                       sep='\t')
+    p_df = p_df.T.drop_duplicates().T
+    p_df2 = pd.read_csv(op.join(dset_dir, 'participants.tsv'),
+                        sep='\t')
+    p_df2 = p_df2.T.drop_duplicates().T
+
+    # Check if row already in participants file
+    matches = p_df[(p_df == p_df2.loc[0]).all(axis=1)]
+    match = matches.index.values.size
+    if not match:
+        p_df = pd.concat((p_df, p_df2))
+        p_df.to_csv(op.join(dset_dir, 'participants.tsv'),
+                    sep='\t', index=False)
+    else:
+        print('Subject/session already found in participants.tsv')
+
+    scratch_sub_dir = op.join(in_dir, 'sub-{0}'.format(sub))
+    out_sub_dir = op.join(dset_dir, 'sub-{0}'.format(sub))
+    if not op.isdir(out_sub_dir):
+        shutil.copytree(scratch_sub_dir, out_sub_dir)
+    elif ses is not None:
+        scratch_ses_dir = op.join(scratch_sub_dir,
+                                  'ses-{0}'.format(ses))
+        out_ses_dir = op.join(out_sub_dir, 'ses-{0}'.format(ses))
+        if not op.isdir(out_ses_dir):
+            shutil.copytree(scratch_ses_dir, out_ses_dir)
+        else:
+            print('Warning: Subject/session directory already exists in '
+                  'dataset.')
+    else:
+        print('Warning: Subject directory already exists in dataset.')