Implemented an option for scan exclusion.

rabies/analysis_pkg/main_wf.py

@@ -26,8 +26,9 @@ def init_main_analysis_wf(preprocess_opts, cr_opts, analysis_opts):
             """)
 
     # filter inclusion/exclusion lists
-    from rabies.utils import filter_scan_inclusion
+    from rabies.utils import filter_scan_inclusion, filter_scan_exclusion
     split_name_list = filter_scan_inclusion(analysis_opts.inclusion_ids, split_name_list)
+    split_name_list = filter_scan_exclusion(analysis_opts.exclusion_ids, split_name_list)
 
     # setting up iterables from the BOLD scan splits
     main_split = pe.Node(niu.IdentityInterface(fields=['split_name']),

rabies/confound_correction_pkg/main_wf.py

@@ -24,8 +24,9 @@ def init_main_confound_correction_wf(preprocess_opts, cr_opts):
         split_dict, split_name, target_list = read_preproc_workflow(preproc_output, nativespace=cr_opts.nativespace_analysis)
 
     # filter inclusion/exclusion lists
-    from rabies.utils import filter_scan_inclusion
+    from rabies.utils import filter_scan_inclusion, filter_scan_exclusion
     split_name = filter_scan_inclusion(cr_opts.inclusion_ids, split_name)
+    split_name = filter_scan_exclusion(cr_opts.exclusion_ids, split_name)
 
     # setting up iterables from the BOLD scan splits
     main_split = pe.Node(niu.IdentityInterface(fields=['split_name']),

rabies/parser.py

@@ -96,6 +96,17 @@ def get_parser():
             "(default: %(default)s)\n"
             "\n"
         )
+    g_execution.add_argument(
+        '--exclusion_ids', type=str,
+        nargs="*",  # 0 or more values expected => creates a list
+        default=['none'],
+        help=
+            "Instead of providing a list of scans to include, this argument provides a list of scans to exclude (while \n"
+            "keeping all other scans). This argument follows the same syntax rules as --includion_ids. --exclusion_ids \n"
+            "and --inclusion_ids cannot be used simultaneously. \n"
+            "(default: %(default)s)\n"
+            "\n"
+        )
     g_execution.add_argument(
         "-p", "--plugin", default='Linear',
         choices=['Linear', 'MultiProc', 'SGE', 'SGEGraph',

rabies/preprocess_pkg/main_wf.py

@@ -135,7 +135,7 @@ def init_main_wf(data_dir_path, output_folder, opts, name='main_wf'):
     bids.config.set_option('extension_initial_dot', True)
     layout = bids.layout.BIDSLayout(data_dir_path, validate=False)
     split_name, scan_info, run_iter, scan_list, bold_scan_list = prep_bids_iter(
-        layout, opts.bold_only, inclusion_list=opts.inclusion_ids)
+        layout, opts.bold_only, inclusion_list=opts.inclusion_ids, exclusion_list=opts.exclusion_ids)
 
     # setting up all iterables
     main_split = pe.Node(niu.IdentityInterface(fields=['split_name', 'scan_info']),

rabies/preprocess_pkg/utils.py

@@ -8,7 +8,7 @@
 )
 from rabies.utils import run_command
 
-def prep_bids_iter(layout, bold_only=False, inclusion_list=['all']):
+def prep_bids_iter(layout, bold_only=False, inclusion_list=['all'], exclusion_list=['none']):
     '''
     This function takes as input a BIDSLayout, and generates iteration lists
     for managing the workflow's iterables depending on whether --bold_only is
@@ -39,9 +39,10 @@ def prep_bids_iter(layout, bold_only=False, inclusion_list=['all']):
             "No functional file with the suffix 'bold' were found among the BIDS directory.")
 
     # filter inclusion/exclusion lists
-    from rabies.utils import filter_scan_inclusion
+    from rabies.utils import filter_scan_inclusion, filter_scan_exclusion
     boldname_list=[pathlib.Path(bold.filename).name.rsplit(".nii")[0] for bold in bold_bids]
     updated_split_name = filter_scan_inclusion(inclusion_list, boldname_list)
+    updated_split_name = filter_scan_exclusion(exclusion_list, updated_split_name)
 
     filtered_bold_bids=[]
     for name in updated_split_name:

rabies/run_main.py

@@ -40,6 +40,13 @@ def execute_workflow(args=None):
         args += input
     log.info(args)
 
+    # inclusion/exclusion list are incompatible parameters
+    if (not opts.inclusion_ids[0]=='all') and (not opts.exclusion_ids[0]=='none'):
+        raise ValueError(f"""
+           Either an inclusion list (--inclusion_ids) or exclusion list (--exclusion_ids)
+           can be provided, not both.
+           """)
+
     if opts.rabies_stage == 'preprocess':
         workflow = preprocess(opts, log)
     elif opts.rabies_stage == 'confound_correction':

rabies/utils.py

@@ -389,6 +389,39 @@ def flatten_list(l):
         return l
 
 
+def filter_scan_exclusion(exclusion_list, split_name):
+    # the function removes a list of scan IDs from split_name
+
+    # exclusion_list: the input provided by the user
+    # split_name: a list of all scan IDs that were found
+
+    import numpy as np
+    import pandas as pd
+    if os.path.isfile(os.path.abspath(exclusion_list[0])):
+        updated_split_name=[]
+        if not '.nii' in pathlib.Path(exclusion_list[0]).name:
+            # read the file as a .txt
+            exclusion_list = np.array(pd.read_csv(os.path.abspath(exclusion_list[0]), header=None)).flatten()
+        for split in split_name:
+            exclude = False
+            for scan in exclusion_list:
+                if split in scan:
+                    exclude = True
+            if not exclude:
+                updated_split_name.append(split)
+    elif exclusion_list[0]=='none':
+        updated_split_name = split_name
+    else:
+        raise ValueError(f"The --exclusion_ids {exclusion_list} input had improper format. It must the full path to a .txt or .nii files.")
+
+    if len(updated_split_name)==0:
+        raise ValueError(f"""
+            No scans are left after scan exclusion!
+            """)
+
+    return updated_split_name
+
+
 def filter_scan_inclusion(inclusion_list, split_name):
     # the function will update the list of scan IDs in split_name to correspond to inclusion/exclusion list