Add some metadata

README.md

@@ -94,15 +94,37 @@ conversion scripts and datainterfaces.  As a placeholder, here we have `src/schn
 
 ## Running a Conversion
 
-To convert the 4 example sessions, simply run
-```bash
-python src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_session.py
-```
-
-To convert the whole dataset, simply run
-```bash
-python src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_dataset.py
-```
+To convert the 5 example sessions,
+1. In `src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_session.py`, update the `data_dir_path` and
+    `output_dir_path` to appropriate local paths. `data_dir_path` should be the high-level directory where the data is
+    stored, corresponding to `Grant Zempolich Project Data` in the GDrive. `output_dir_path` can be any valid path on
+    your system where the output NWB files will be stored.
+2. simply run
+    ```bash
+    python src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_session.py
+    ```
+
+To convert the whole dataset,
+1. Update `data_dir_path` and `output_dir_path` in `src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_all_sessions.py`
+    as with the example sessions.
+2. simply run
+    ```bash
+    python src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_all_sessions.py
+    ```
 
 Note that the dataset conversion uses multiprocessing, currently set to 4 workers.  To use more or fewer workers, simply
 change the `max_workers` argument to `dataset_to_nwb()`.
+
+### TODOs
+
+There are some placeholders in the current version of the conversion that will need to be filled in by the Schneider Lab
+before the conversion can be completed with the full data/metadata. These placeholders are marked with TODOs in the code
+to make them easier to spot, and a list is provided below for convenience:
+
+* In `src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_open_ephys_recording_interface.py` Line 36,
+    channel_positions are truncated to account for the 1-channel ephys data provided in the google drive. Lines 36-37
+    will need to be removed to enable running the conversion on the full ephys data.
+* In `src/schneider_lab_to_nwb/zempolich_2024/metadata.yaml` Line 29, the mapping between subject_id and genotype is a
+    placeholder. Please specify the genotype for each subject, and it will automatically propagate to the NWB file.
+* In `src/schneider_lab_to_nwb/zempolich_2024/metadata.yaml` Line 51, the mapping between subject_id and sex is a
+    placeholder. Please specify the sex for each subject, and it will automatically propagate to the NWB file.

src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_convert_session.py

@@ -105,6 +105,12 @@ def session_to_nwb(
     editable_metadata = load_dict_from_file(editable_metadata_path)
     metadata = dict_deep_update(metadata, editable_metadata)
 
+    # Add session description to metadata
+    behavior_folder_name = behavior_file_path.parent.name
+    session_metadata = next(meta for meta in metadata["Session"] if meta["name"] == behavior_folder_name)
+    session_description = session_metadata["description"]
+    metadata["NWBFile"]["session_description"] = session_description
+
     add_session_start_time_to_metadata(
         behavior_file_path=behavior_file_path, ephys_folder_path=ephys_folder_path, metadata=metadata
     )
@@ -116,13 +122,22 @@ def session_to_nwb(
     for i, video_file_path in enumerate(video_file_paths):
         metadata_key_name = f"VideoCamera{i+1}"
         metadata["Behavior"][metadata_key_name] = editable_metadata["Behavior"][metadata_key_name]
+    # remove extra videos from metadata
+    i += 1
+    while f"VideoCamera{i+1}" in metadata["Behavior"]:
+        metadata["Behavior"].pop(f"VideoCamera{i+1}")
+        i += 1
 
     subject_id = behavior_file_path.name.split("_")[1]
     session_id = behavior_file_path.name.split("_")[2]
     nwbfile_path = output_dir_path / f"sub-{subject_id}_ses-{session_id}.nwb"
     metadata["NWBFile"]["session_id"] = session_id
     metadata["Subject"]["subject_id"] = subject_id
 
+    # Add subject info to metadata
+    metadata["Subject"]["sex"] = metadata["SubjectMaps"]["subject_id_to_sex"][subject_id]
+    metadata["Subject"]["genotype"] = metadata["SubjectMaps"]["subject_id_to_genotype"][subject_id]
+
     # Run conversion
     converter.run_conversion(metadata=metadata, nwbfile_path=nwbfile_path, conversion_options=conversion_options)
 
@@ -227,6 +242,20 @@ def main():
         verbose=verbose,
     )
 
+    # Example Session missing Camera 2
+    behavior_file_path = data_dir_path / "M2_EphysBehavioralFiles" / "raw_m80_240819_001.mat"
+    video_folder_path = data_dir_path / "Videos" / "M2EphysVideos" / "m80" / "240819"
+    intrinsic_signal_optical_imaging_folder_path = data_dir_path / "Intrinsic Imaging Data" / "m80"
+    session_to_nwb(
+        behavior_file_path=behavior_file_path,
+        video_folder_path=video_folder_path,
+        intrinsic_signal_optical_imaging_folder_path=intrinsic_signal_optical_imaging_folder_path,
+        brain_region="M2",
+        output_dir_path=output_dir_path,
+        stub_test=stub_test,
+        verbose=verbose,
+    )
+
 
 if __name__ == "__main__":
     main()

src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_metadata.yaml

@@ -10,10 +10,10 @@ NWBFile:
   experimenter:
     - Zempolich, Grant W.
     - Schneider, David M.
+
 Subject:
   species: Mus musculus
   age: P12W/  # in ISO 8601, such as "P1W2D"
-  sex: U  # One of M, F, U, or O
   description: 12-week-old C57BL/6 or VGATChR2-EYFP mice were used for all behavioral, electrophysiology, and optogenetic experiments. In the VGAT-ChR2-EYFP mouse line, channelrhodopsin (ChR2) was coupled to the vesicular GABA transporter, inducing expression in GABAergic inhibitory neurons ubiquitously across cortex and allowing for real time optogenetic inhibition of brain regions of interest.
   strain: C57BL/6
 
@@ -25,6 +25,62 @@ BrainRegion:
     electrode_group_location: Secondary Motor Cortex (M2)
     optogenetic_stimulus_site_location: Secondary Motor Cortex (1.0-1.5 AP, 0.5-0.7 ML)
 
+SubjectMaps:
+  subject_id_to_genotype: # TODO: Replace with actual genotype
+    m53: C57BL/6 or VGATChR2-EYFP
+    m54: C57BL/6 or VGATChR2-EYFP
+    m55: C57BL/6 or VGATChR2-EYFP
+    m63: C57BL/6 or VGATChR2-EYFP
+    m64: C57BL/6 or VGATChR2-EYFP
+    m65: C57BL/6 or VGATChR2-EYFP
+    m66: C57BL/6 or VGATChR2-EYFP
+    m67: C57BL/6 or VGATChR2-EYFP
+    m69: C57BL/6 or VGATChR2-EYFP
+    m70: C57BL/6 or VGATChR2-EYFP
+    m71: C57BL/6 or VGATChR2-EYFP
+    m72: C57BL/6 or VGATChR2-EYFP
+    m73: C57BL/6 or VGATChR2-EYFP
+    m74: C57BL/6 or VGATChR2-EYFP
+    m75: C57BL/6 or VGATChR2-EYFP
+    m76: C57BL/6 or VGATChR2-EYFP
+    m77: C57BL/6 or VGATChR2-EYFP
+    m78: C57BL/6 or VGATChR2-EYFP
+    m79: C57BL/6 or VGATChR2-EYFP
+    m80: C57BL/6 or VGATChR2-EYFP
+    m81: C57BL/6 or VGATChR2-EYFP
+  subject_id_to_sex: # TODO: Replace with actual sex (M, F, U, or O)
+    m53: U
+    m54: U
+    m55: U
+    m63: U
+    m64: U
+    m65: U
+    m66: U
+    m67: U
+    m69: U
+    m70: U
+    m71: U
+    m72: U
+    m73: U
+    m74: U
+    m75: U
+    m76: U
+    m77: U
+    m78: U
+    m79: U
+    m80: U
+    m81: U
+
+Session:
+  - name: A1_EphysBehavioralFiles
+    description: Mice performed the auditory guided task while electricophysiological neural activity was recorded in the primary auditory cortex (A1).
+  - name: A1_OptoBehavioralFiles
+    description: Mice performed the auditory guided task while optogenetic stimulation was delivered to the primary auditory cortex (A1).
+  - name: M2_EphysBehavioralFiles
+    description: Mice performed the auditory guided task while electricophysiological neural activity was recorded in the secondary motor cortex (M2).
+  - name: M2_OptoBehavioralFiles
+    description: Mice performed the auditory guided task while optogenetic stimulation was delivered to the secondary motor cortex (M2).
+
 Ecephys:
   Device:
     - name: MasmanidisSiliconMicroprobe128AxN
@@ -37,7 +93,7 @@ Ecephys:
   ElectricalSeries:
     - name: ElectricalSeries
       description: Recording of AC neural responses in mice performing this behavioral task utilized dense 128-channel recording probes (Masmanidis Lab). These recording probes span a depth ~1mm, allowing for sampling of all layers of cortex. Electrophysiology data was recorded using OpenEphys Acquisition Board v2.4 and associated OpenEphys GUI software.
-  folder_name_to_start_datetime:
+  folder_name_to_start_datetime: # This mapping is used for ambiguous start times from OpenEphys ex. '18413' could be 18:41:03 or 18:04:13
     m53/Day1_A1: 2023-10-29T16:56:01
     m54/Day1_A1: 2023-10-29T18:18:03
     m65/Day2_A1: 2023-10-26T18:49:04

src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_notes.md

@@ -31,8 +31,6 @@
 
 ## Active Requests
 - Mice sexes
-- target image for m73
-- Cam2 Video for M2EphysVideos/m80/240819
 
 ## Questions for Midway Meeting
 - injection vs stimulation location(s) for A1 vs M2???

src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_nwbconverter.py

@@ -44,8 +44,10 @@ def temporally_align_data_interfaces(self) -> None:
             starting_timestamp = get_starting_timestamp(mat_file=file)
             cam1_timestamps -= starting_timestamp
             cam2_timestamps -= starting_timestamp
-        self.data_interface_objects["VideoCamera1"].set_aligned_timestamps([cam1_timestamps])
-        self.data_interface_objects["VideoCamera2"].set_aligned_timestamps([cam2_timestamps])
+        if "VideoCamera1" in self.data_interface_objects:
+            self.data_interface_objects["VideoCamera1"].set_aligned_timestamps([cam1_timestamps])
+        if "VideoCamera2" in self.data_interface_objects:
+            self.data_interface_objects["VideoCamera2"].set_aligned_timestamps([cam2_timestamps])
 
     # NOTE: passing in conversion_options as an attribute is a temporary solution until the neuroconv library is updated
     #  to allow for easier customization of the conversion process

src/schneider_lab_to_nwb/zempolich_2024/zempolich_2024_open_ephys_recording_interface.py

@@ -33,7 +33,7 @@ def add_to_nwbfile(
         """
         folder_path = self.source_data["folder_path"]
         channel_positions = np.load(folder_path / "channel_positions.npy")
-        if True:  # TODO: Replace with `if stub_test:` once all channels are present in the data
+        if True:  # TODO: Remove this if block once all channels are present in the data
             channel_positions = channel_positions[:1, :]
         location = metadata["BrainRegion"][brain_region]["electrode_group_location"]
         for electrode_group in metadata["Ecephys"]["ElectrodeGroup"]: