[BUG] Prepend sub- prefix if it is not in participants.tsv

src/brainflex/core/get_data.py

@@ -181,6 +181,8 @@ def _get_freesurfer(
 def _write_pheno(layout: DatasetLayout, session_ids: dict[str, str]) -> None:
     """Copy participants.tsv to the tabular directory, with a session ID column.
 
+    Participant IDs without the ``sub-`` prefix have it prepended.
+
     Rows with an empty participant ID and participants without BIDS imaging
     data are excluded. Warns if participant IDs and BIDS subjects do not fully match,
     and raises ValueError if none match.
@@ -196,6 +198,16 @@ def _write_pheno(layout: DatasetLayout, session_ids: dict[str, str]) -> None:
         fieldnames = reader.fieldnames
         rows = [row for row in reader if (row[COL_PARTICIPANT_ID] or "").strip()]
 
+    for row in rows:
+        participant_id = row[COL_PARTICIPANT_ID]
+        if not participant_id.startswith(BIDS_SUBJECT_PREFIX):
+            row[COL_PARTICIPANT_ID] = BIDS_SUBJECT_PREFIX + participant_id
+            warnings.warn(
+                f"Prepended {BIDS_SUBJECT_PREFIX!r} to participant ID"
+                f" {participant_id!r} in {PARTICIPANTS_FILENAME}",
+                stacklevel=3,
+            )
+
     fieldnames = [COL_PARTICIPANT_ID, COL_SESSION_ID] + [
         col for col in fieldnames if col not in (COL_PARTICIPANT_ID, COL_SESSION_ID)
     ]

tests/core/test_get_data.py

@@ -204,11 +204,32 @@ def test_no_matching_participants(
         self, mock_s3: dict[tuple[str, str], bytes], root: Path
     ):
         mock_s3[(OPENNEURO_BUCKET, f"{DATASET_ID}/participants.tsv")] = (
-            b"participant_id\tage\nSZ_0113\t25\n"
+            b"participant_id\tage\nsub-SZ0113\t25\n"
         )
         with pytest.raises(ValueError, match="No participant IDs"):
             get_data(root, DATASET_ID)
 
+    def test_participants_without_prefix(
+        self, mock_s3: dict[tuple[str, str], bytes], root: Path
+    ):
+        mock_s3[(OPENNEURO_BUCKET, f"{DATASET_ID}/participants.tsv")] = (
+            b"participant_id\tage\n01\t20\n02\t30\nsub-03\t40\n"
+        )
+        with pytest.warns(UserWarning) as record_warnings:
+            dpath_study = get_data(root, DATASET_ID)
+        assert [
+            str(w.message) for w in record_warnings if "Prepended" in str(w.message)
+        ] == [
+            "Prepended 'sub-' to participant ID '01' in participants.tsv",
+            "Prepended 'sub-' to participant ID '02' in participants.tsv",
+        ]
+        pheno = (dpath_study / "tabular" / "pheno.tsv").read_text()
+        assert pheno == (
+            "participant_id\tsession_id\tage\n"
+            "sub-01\tses-unnamed\t20\n"
+            "sub-02\tses-unnamed\t30\n"
+        )
+
     def test_no_participants_json(
         self, mock_s3: dict[tuple[str, str], bytes], root: Path
     ):