bidsify output

giga_connectome/postprocess.py

@@ -74,8 +74,9 @@ def run_postprocessing_dataset(
         Smoothing kernel size, passed to nilearn masker.
 
     output_path:
-        Full path to output file, named in the following format:
-        output_dir / atlas-<atlas>_desc-<strategy_name>.h5
+        Full path to output directory.
+        Will be used to name the output file in the following format:
+        [sub-<sub>][_ses-<ses>_]atlas-<atlas>_desc-<strategy_name>_relmat.h5
 
     analysis_level : str
         Level of analysis, either "participant" or "group".
@@ -105,6 +106,16 @@ def run_postprocessing_dataset(
         )
         # parse file name
         subject, session, specifier = utils.parse_bids_name(img.path)
+
+        connectome_path = output_path / subject
+        if session:
+            connectome_path = connectome_path / session
+        filename = utils.output_filename(
+            Path(img.filename).stem, atlas, strategy["name"]
+        )
+        connectome_path = connectome_path / "func" / filename
+        connectome_path = utils.check_path(connectome_path, verbose=True)
+
         for desc, masker in atlas_maskers.items():
             attribute_name = f"{subject}_{specifier}_atlas-{atlas}_desc-{desc}"
             if not denoised_img:
@@ -126,8 +137,8 @@ def run_postprocessing_dataset(
             connectomes[desc].append(correlation_matrix)
 
             # dump to h5
-            flag = _set_file_flag(output_path)
-            with h5py.File(output_path, flag) as f:
+            flag = _set_file_flag(connectome_path)
+            with h5py.File(connectome_path, flag) as f:
                 group = _fetch_h5_group(f, subject, session)
                 timeseries_dset = group.create_dataset(
                     f"{attribute_name}_timeseries", data=time_series_atlas
@@ -141,11 +152,18 @@ def run_postprocessing_dataset(
 
     if analysis_level == "group":
         print("create group connectome")
+        connectome_path = (
+            output_path
+            / "group"
+            / utils.output_filename("", atlas, strategy["name"])
+        )
+        connectome_path = utils.check_path(connectome_path, verbose=True)
+        print(connectome_path)
         for desc in connectomes:
             average_connectome = np.mean(
                 np.array(connectomes[desc]), axis=0
             ).astype(np.float32)
-            with h5py.File(output_path, "a") as f:
+            with h5py.File(connectome_path, "a") as f:
                 f.create_dataset(
                     f"atlas-{atlas}_desc-{desc}_connectome",
                     data=average_connectome,

giga_connectome/tests/test_cli.py

@@ -31,7 +31,6 @@ def test_help(capsys):
 
 @pytest.mark.smoke
 def test_smoke(tmp_path, capsys):
-
     bids_dir = resource_filename(
         "giga_connectome",
         "data/test_data/ds000017-fmriprep22.0.1-downsampled-nosurface",

giga_connectome/utils.py

@@ -1,9 +1,15 @@
-from typing import List, Tuple, Union
+from __future__ import annotations
+
+from typing import List, Tuple, Union, Any
+import json
 from pathlib import Path
+
 from nilearn.interfaces.bids import parse_bids_filename
 from bids.layout import Query
 from bids import BIDSLayout
 
+from giga_connectome import __version__
+
 
 def get_bids_images(
     subjects: List[str],
@@ -184,6 +190,8 @@ def check_path(path: Path, verbose=True):
     ext = path.suffix
     path_parent = path.parent
 
+    path_parent.mkdir(parents=True, exist_ok=True)
+
     if path.exists():
         similar_paths = [
             str(p).replace(ext, "")
@@ -199,3 +207,60 @@ def check_path(path: Path, verbose=True):
         if verbose:
             print(f"Specified path already exists, using {path} instead.")
     return path
+
+
+def create_ds_description(output_dir: Path) -> None:
+    """Create a dataset_description.json file."""
+    ds_desc: dict[str, Any] = {
+        "BIDSVersion": "1.9.0",
+        "License": None,
+        "Name": None,
+        "ReferencesAndLinks": [],
+        "Authors": [
+            "Foo",
+            "Bar",
+        ],
+        "DatasetDOI": None,
+        "DatasetType": "derivative",
+        "GeneratedBy": [
+            {
+                "Name": "giga_connectome",
+                "Version": __version__,
+                "CodeURL": "https://github.com/SIMEXP/giga_connectome.git",
+            }
+        ],
+        "HowToAcknowledge": (
+            "Please refer to our repository: "
+            "https://github.com/SIMEXP/giga_connectome.git."
+        ),
+    }
+    with open(output_dir / "dataset_description.json", "w") as f:
+        json.dump(ds_desc, f, indent=4)
+
+
+def create_sidecar(output_path: Path) -> None:
+    """Create a JSON sidecar for the connectivity data."""
+    metadata: dict[str, Any] = {
+        "NodeFiles": "REQUIRED",
+        "Measure": "Pearson correlation",
+        "MeasureDescription": "Pearson correlation",
+        "Weighted": "REQUIRED",
+        "Directed": False,
+        "ValidDiagonal": True,
+        "StorageFormat": "Full",
+        "NonNegative": "",
+        "Code": "https://github.com/SIMEXP/giga_connectome.git",
+    }
+    with open(output_path, "w") as f:
+        json.dump(metadata, f, indent=4)
+
+
+def output_filename(source_file: str, atlas: str, strategy: str) -> str:
+    root = source_file.split("_")[:-1]
+    root = "_".join(root)
+    if len(root) > 0:
+        root += "_"
+    return (
+        f"{root}atlas-{atlas}_meas-PearsonCorrelation_desc-{strategy}"
+        "_relmat.h5"
+    )

giga_connectome/workflow.py

@@ -40,6 +40,12 @@ def workflow(args):
     )
     print("Indexing BIDS directory")
 
+    utils.create_ds_description(output_dir)
+    utils.create_sidecar(
+        output_dir
+        / f"meas-PearsonCorrelation_desc-{args.denoise_strategy}_relmat.json"
+    )
+
     # create subject ts and connectomes
     # refactor the two cases into one
 
@@ -51,11 +57,7 @@ def workflow(args):
             group_mask, resampled_atlases = generate_gm_mask_atlas(
                 working_dir, atlas, template, subj_data["mask"]
             )
-            connectome_path = output_dir / (
-                f"sub-{subject}_atlas-{atlas['name']}"
-                f"_desc-{strategy['name']}.h5"
-            )
-            connectome_path = utils.check_path(connectome_path, verbose=True)
+
             print("Generate subject level connectomes")
             run_postprocessing_dataset(
                 strategy,
@@ -64,7 +66,7 @@ def workflow(args):
                 group_mask,
                 standardize,
                 smoothing_fwhm,
-                connectome_path,
+                output_dir,
                 analysis_level,
                 calculate_average_correlation,
             )
@@ -77,11 +79,6 @@ def workflow(args):
     group_mask, resampled_atlases = generate_gm_mask_atlas(
         working_dir, atlas, template, subj_data["mask"]
     )
-    connectome_path = (
-        output_dir / f"atlas-{atlas['name']}_desc-{strategy['name']}.h5"
-    )
-    connectome_path = utils.check_path(connectome_path, verbose=True)
-    print(connectome_path)
     print("Generate subject level connectomes")
     run_postprocessing_dataset(
         strategy,
@@ -90,7 +87,7 @@ def workflow(args):
         group_mask,
         standardize,
         smoothing_fwhm,
-        connectome_path,
+        output_dir,
         analysis_level,
         calculate_average_correlation,
     )