docs/changes/newsfragments/183.bugfix

@@ -0,0 +1 @@
+Fix a bug in which only ``REST1`` and ``REST2`` tasks could be accesed in :class:`.DataladHCP1200` and :class:`.HCP1200` datagrabbers by `Fede Raimondo`_

docs/changes/newsfragments/183.change

@@ -0,0 +1 @@
+Add ``ica_fix`` parameter to :class:`.DataladHCP1200` and :class:`.HCP1200` datagrabbers to allow for selecting data processed with ICA+FIX. Default value is ``False`` which changes behaviour since 0.0.1 release. By `Fede Raimondo`_

junifer/datagrabber/hcp.py

@@ -26,9 +26,12 @@ class HCP1200(PatternDataGrabber):
     phase_encodings : {"LR", "RL"} or list of the options, optional
         HCP phase encoding directions. If None, both will be used
         (default None).
+    ica_fix : bool, optional
+        Whether to retrieve data that was processed with ICA+FIX.
+        Only 'REST1' and 'REST2' tasks are available with ICA+FIX (default
+        False).
     **kwargs
         Keyword arguments passed to superclass.
-
     """
 
     def __init__(
@@ -36,6 +39,7 @@ class HCP1200(PatternDataGrabber):
         datadir: Union[str, Path],
         tasks: Union[str, List[str], None] = None,
         phase_encodings: Union[str, List[str], None] = None,
+        ica_fix: bool = False,
         **kwargs,
     ) -> None:
         # All tasks
@@ -82,6 +86,12 @@ class HCP1200(PatternDataGrabber):
                     f"{all_phase_encodings}."
                 )
 
+        if ica_fix:
+            if not all([task in ["REST1", "REST2"] for task in self.tasks]):
+                raise_error(
+                    "ICA+FIX is only available for 'REST1' and 'REST2' tasks."
+                )
+        suffix = "_hp2000_clean" if ica_fix else ""
         # The types of data
         types = ["BOLD"]
         # The patterns
@@ -89,7 +99,8 @@ class HCP1200(PatternDataGrabber):
             "BOLD": (
                 "{subject}/MNINonLinear/Results/"
                 "{task}_{phase_encoding}/"
-                "{task}_{phase_encoding}_hp2000_clean.nii.gz"
+                "{task}_{phase_encoding}"
+                f"{suffix}.nii.gz"
             )
         }
         # The replacements
@@ -173,6 +184,10 @@ class DataladHCP1200(DataladDataGrabber, HCP1200):
     phase_encodings : {"LR", "RL"} or list of the options, optional
         HCP phase encoding directions. If None, both will be used
         (default None).
+    ica_fix : bool, optional
+        Whether to retrieve data that was processed with ICA+FIX.
+        Only 'REST1' and 'REST2' tasks are available with ICA+FIX (default
+        False).
     """
 
     def __init__(
@@ -180,6 +195,7 @@ class DataladHCP1200(DataladDataGrabber, HCP1200):
         datadir: Union[str, Path, None] = None,
         tasks: Union[str, List[str], None] = None,
         phase_encodings: Union[str, List[str], None] = None,
+        ica_fix: bool = False,
     ) -> None:
         uri = (
             "https://github.com/datalad-datasets/"
@@ -192,6 +208,7 @@ class DataladHCP1200(DataladDataGrabber, HCP1200):
             phase_encodings=phase_encodings,
             uri=uri,
             rootdir=rootdir,
+            ica_fix=ica_fix,
         )
 
     @property

junifer/datagrabber/tests/test_hcp.py

@@ -29,33 +29,38 @@ def hcpdg() -> Iterable[DataladHCP1200]:
 
 
 @pytest.mark.parametrize(
-    "tasks, phase_encodings, expected_path_name",
+    "tasks, phase_encodings, ica_fix, expected_path_name",
     [
-        (None, None, "rfMRI_REST1_LR_hp2000_clean.nii.gz"),
+        (None, None, False, "rfMRI_REST1_LR.nii.gz"),
-        ("REST1", "LR", "rfMRI_REST1_LR_hp2000_clean.nii.gz"),
+        ("REST1", "LR", False, "rfMRI_REST1_LR.nii.gz"),
-        ("REST1", "RL", "rfMRI_REST1_RL_hp2000_clean.nii.gz"),
+        ("REST1", "RL", False, "rfMRI_REST1_RL.nii.gz"),
-        ("REST2", "LR", "rfMRI_REST2_LR_hp2000_clean.nii.gz"),
+        ("REST2", "LR", False, "rfMRI_REST2_LR.nii.gz"),
-        ("REST2", "RL", "rfMRI_REST2_RL_hp2000_clean.nii.gz"),
+        ("REST2", "RL", False, "rfMRI_REST2_RL.nii.gz"),
-        ("SOCIAL", "LR", "tfMRI_SOCIAL_LR_hp2000_clean.nii.gz"),
+        ("SOCIAL", "LR", False, "tfMRI_SOCIAL_LR.nii.gz"),
-        ("SOCIAL", "RL", "tfMRI_SOCIAL_RL_hp2000_clean.nii.gz"),
+        ("SOCIAL", "RL", False, "tfMRI_SOCIAL_RL.nii.gz"),
-        ("WM", "LR", "tfMRI_WM_LR_hp2000_clean.nii.gz"),
+        ("WM", "LR", False, "tfMRI_WM_LR.nii.gz"),
-        ("WM", "RL", "tfMRI_WM_RL_hp2000_clean.nii.gz"),
+        ("WM", "RL", False, "tfMRI_WM_RL.nii.gz"),
-        ("RELATIONAL", "LR", "tfMRI_RELATIONAL_LR_hp2000_clean.nii.gz"),
+        ("RELATIONAL", "LR", False, "tfMRI_RELATIONAL_LR.nii.gz"),
-        ("RELATIONAL", "RL", "tfMRI_RELATIONAL_RL_hp2000_clean.nii.gz"),
+        ("RELATIONAL", "RL", False, "tfMRI_RELATIONAL_RL.nii.gz"),
-        ("EMOTION", "LR", "tfMRI_EMOTION_LR_hp2000_clean.nii.gz"),
+        ("EMOTION", "LR", False, "tfMRI_EMOTION_LR.nii.gz"),
-        ("EMOTION", "RL", "tfMRI_EMOTION_RL_hp2000_clean.nii.gz"),
+        ("EMOTION", "RL", False, "tfMRI_EMOTION_RL.nii.gz"),
-        ("LANGUAGE", "LR", "tfMRI_LANGUAGE_LR_hp2000_clean.nii.gz"),
+        ("LANGUAGE", "LR", False, "tfMRI_LANGUAGE_LR.nii.gz"),
-        ("LANGUAGE", "RL", "tfMRI_LANGUAGE_RL_hp2000_clean.nii.gz"),
+        ("LANGUAGE", "RL", False, "tfMRI_LANGUAGE_RL.nii.gz"),
-        ("GAMBLING", "LR", "tfMRI_GAMBLING_LR_hp2000_clean.nii.gz"),
+        ("GAMBLING", "LR", False, "tfMRI_GAMBLING_LR.nii.gz"),
-        ("GAMBLING", "RL", "tfMRI_GAMBLING_RL_hp2000_clean.nii.gz"),
+        ("GAMBLING", "RL", False, "tfMRI_GAMBLING_RL.nii.gz"),
-        ("MOTOR", "LR", "tfMRI_MOTOR_LR_hp2000_clean.nii.gz"),
+        ("MOTOR", "LR", False, "tfMRI_MOTOR_LR.nii.gz"),
-        ("MOTOR", "RL", "tfMRI_MOTOR_RL_hp2000_clean.nii.gz"),
+        ("MOTOR", "RL", False, "tfMRI_MOTOR_RL.nii.gz"),
+        ("REST1", "LR", True, "rfMRI_REST1_LR_hp2000_clean.nii.gz"),
+        ("REST1", "RL", True, "rfMRI_REST1_RL_hp2000_clean.nii.gz"),
+        ("REST2", "LR", True, "rfMRI_REST2_LR_hp2000_clean.nii.gz"),
+        ("REST2", "RL", True, "rfMRI_REST2_RL_hp2000_clean.nii.gz"),
     ],
 )
 def test_hcp1200_datagrabber(
     hcpdg: DataladHCP1200,
     tasks: Optional[str],
     phase_encodings: Optional[str],
+    ica_fix: bool,
     expected_path_name: str,
 ) -> None:
     """Test HCP1200 datagrabber.
@@ -69,6 +74,8 @@ def test_hcp1200_datagrabber(
         The parametrized tasks.
     phase_encodings : str
         The parametrized phase encodings.
+    ica_fix : bool
+        The parametrized ICA-FIX flag.
     expected_path_name : str
         The parametrized expected path name.
 
@@ -78,6 +85,7 @@ def test_hcp1200_datagrabber(
         datadir=hcpdg.datadir,
         tasks=tasks,
         phase_encodings=phase_encodings,
+        ica_fix=ica_fix,
     )
     # Get all elements
     all_elements = dg.get_elements()
@@ -342,3 +350,38 @@ def test_hcp1200_datagrabber_elements(
             assert element[2] in ["LR", "RL"]
 
         assert set(found_subjects) == set(expected_subjects)
+
+
+@pytest.mark.parametrize(
+    "tasks, ica_fix",
+    [
+        ("SOCIAL", True),
+        ("WM", True),
+        ("RELATIONAL", True),
+        ("EMOTION", True),
+        ("LANGUAGE", True),
+        ("GAMBLING", True),
+        ("MOTOR", True),
+    ],
+)
+def test_hcp1200_datagrabber_incorrect_access_icafix(
+    tasks: Optional[str],
+    ica_fix: bool
+) -> None:
+    """Test HCP1200 datagrabber incorrect access for icafix.
+
+    Parameters
+    ----------
+    tasks : str
+        The parametrized tasks.
+    ica_fix : bool
+        The parametrized ICA-FIX flag.
+
+    """
+    configure_logging(level="DEBUG")
+    with pytest.raises(ValueError, match="is only available for"):
+        _ = HCP1200(
+            datadir=".",
+            tasks=tasks,
+            ica_fix=ica_fix,
+        )

tools/create_hcp1200_example_dataset.py

@@ -22,7 +22,7 @@ if __name__ == "__main__":
         # Set base directory
         basedir = tmpdir_path / "example_hcp1200"
         # Create new datalad dataset
-        dataset = dl.create(path=str(basedir.absolute()))
+        dataset = dl.create(path=str(basedir.absolute()))  # type: ignore
         # Generate subject directories
         for sub in range(1, 10):
             subdir = basedir / f"sub-{sub:02d}"
@@ -57,15 +57,26 @@ if __name__ == "__main__":
                 )
                 # Create subject data directory
                 sub_datadir.mkdir(parents=True)
+
                 # Set subject data file
                 sub_datafile = (
                     sub_datadir
-                    / f"{new_task}_{phase_encoding}_hp2000_clean.nii.gz"
+                    / f"{new_task}_{phase_encoding}.nii.gz"
                 )
                 # Create subject data file
                 with open(sub_datafile, "w") as f:
                     f.write("placeholder")
 
+                if "REST" in task:
+                    # Set subject data file with ICA+FIX
+                    sub_datafile = (
+                        sub_datadir
+                        / f"{new_task}_{phase_encoding}_hp2000_clean.nii.gz"
+                    )
+                    # Create subject data file with ICA+FIX
+                    with open(sub_datafile, "w") as f:
+                        f.write("placeholder")
+
         # Save datalad dataset
         dataset.save(recursive=True)
         # Add datalad sibling