script to get ct for juliette

Author: singlesp

misc/get_sch400_ct.py

@@ -0,0 +1,90 @@
+"""
+Collect Schaefer 2018 400-parcel cortical thickness (thickavg) values from
+per-subject FreeSurfer surface stats TSVs and write a wide-format summary TSV.
+
+Input:
+    /cbica/projects/grmpy/data/derivatives/freesurfer-post/sub-*/
+        sub-*_seg-Schaefer2018400Parcels7Networks_surfacestats.tsv
+
+Output:
+    /cbica/projects/grmpy/code/misc/Schaefer2018400Parcels7Networks_ct.tsv
+
+Columns: participant_id | <one column per ROI (thickavg)> | mean_ct
+The Background+FreeSurfer_Defined_Medial_Wall ROI is excluded.
+"""
+
+import glob
+import os
+
+import pandas as pd
+
+DERIV_DIR = "/cbica/projects/grmpy/data/derivatives/freesurfer-post"
+TSV_GLOB = os.path.join(
+    DERIV_DIR, "sub-*", "sub-*_seg-Schaefer2018400Parcels7Networks_surfacestats.tsv"
+)
+OUTPUT_PATH = "/cbica/projects/grmpy/code/misc/Schaefer2018400Parcels7Networks_ct.tsv"
+MEDIAL_WALL = "Background+FreeSurfer_Defined_Medial_Wall"
+
+
+def process_subject(tsv_path: str) -> pd.Series | None:
+    df = pd.read_csv(tsv_path, sep="\t")
+
+    df = df[df["structname"] != MEDIAL_WALL].copy()
+
+    if df.empty:
+        print(f"  WARNING: no ROIs remaining after filtering medial wall in {tsv_path}")
+        return None
+
+    participant_id = df["participant_id"].iloc[0]
+
+    roi_series = df.set_index("structname")["thickavg"]
+    roi_series.name = None
+
+    result = pd.Series({"participant_id": participant_id})
+    result = pd.concat([result, roi_series])
+    result["mean_ct"] = roi_series.mean()
+
+    return result
+
+
+def main():
+    tsv_files = sorted(glob.glob(TSV_GLOB))
+
+    if not tsv_files:
+        raise FileNotFoundError(
+            f"No TSV files found matching pattern:\n  {TSV_GLOB}\n"
+            "Check that the derivatives directory is mounted and the file naming is correct."
+        )
+
+    print(f"Found {len(tsv_files)} subject TSV(s). Processing...")
+
+    rows = []
+    for path in tsv_files:
+        print(f"  {os.path.basename(os.path.dirname(path))}")
+        row = process_subject(path)
+        if row is not None:
+            rows.append(row)
+
+    if not rows:
+        raise RuntimeError("No valid data was collected. Output not written.")
+
+    output_df = pd.DataFrame(rows).reset_index(drop=True)
+
+    cols = (
+        ["participant_id"]
+        + [c for c in output_df.columns if c not in ("participant_id", "mean_ct")]
+        + ["mean_ct"]
+    )
+    output_df = output_df[cols]
+
+    os.makedirs(os.path.dirname(OUTPUT_PATH), exist_ok=True)
+    output_df.to_csv(OUTPUT_PATH, sep="\t", index=False)
+
+    print(
+        f"\nDone. {len(output_df)} participant(s), {len(output_df.columns) - 2} ROI columns."
+    )
+    print(f"Output written to:\n  {OUTPUT_PATH}")
+
+
+if __name__ == "__main__":
+    main()