Analysis GUI

Goal: An analysis gui should be aware of files within a directory of specific file type and with specific filename-endings.

Description: A core component is to allow the exploration of data from an experimental session. A session usually produces two memory mapped ome tiff stacks and wheel encoder speed. The software should synchronize the data to a common time dimension, allowing users to scrub through all of the data at once, synchronously.

This is accomplished by synchronizing the time metadata from several files _encoder-data.csv _meso.ome.tiff_frame_metadata.json _thor.ome.tiff_frame_metadata.json

The file hierarchy is standardized in BIDS format. Users should be able to select a folder where this file structure exists within:

• PsychoPy trial data and wheel encoder data are found in the /beh folder data/protocol/subject/session/beh
• The ome.tiff stacks are found in the /func folder data/protocol/subject/session/func

data/
    protocol/
        subject/
            session/
                func/
                beh/

Current example code

def load_metadata(directory):
    frame_metadata_df = None
    pupil_frame_metadata_df = None
 
    # Parse the directory for files ending with '_frame_metadata.json' and 'pupil_frame_metadata.json'
    for file in os.listdir(directory):
        if file.endswith('meso_frame_metadata.jsonf'): 
            frame_metadata_path = os.path.join(directory, file)
            frame_metadata_df = load_frame_metadata(frame_metadata_path)
        elif file.endswith('pupil_frame_metadata.jsonf'):
            pupil_frame_metadata_path = os.path.join(directory, file)
            pupil_frame_metadata_df = load_frame_metadata(pupil_frame_metadata_path)
 
    return frame_metadata_df, pupil_frame_metadata_df

def plot_encoder_csv(encoder_df, psychopy_df):
    speed = encoder_df.iloc[:, 2]
    time = encoder_df.iloc[:, 1]
 
    plt.plot(time, speed)
    plt.xlabel('Time')
    plt.ylabel('Speed')
 
    for idx, row in psychopy_df.iterrows():
        if pd.notna(row['stim_label']):
            start_time = row['Checkerboard.started']
            label = f"{row['stim_label']}_{row['direction']}"
            plt.axvline(x=start_time, linestyle='--', label=label)
 
    plt.title('Speed over Time')
    plt.legend()
    plt.show()

        dh_md_df, th_md_df = data.load_metadata(self.config.bids_dir)
   data.plot_encoder_csv(data.load_wheel_data(self.config.bids_dir), data.load_psychopy_data(self.config.bids_dir))
 

Then, the file paths for the tiff stacks are standardized as: _meso.ome _pupil.ome

The result should be a gui that shows both the meso and pupil videos above the speed plot for a session. There should be a dropdown selection that parses json files in the parent directory above the BIDS directory; once a file is selected it should be displayed as a table for viewing.

File Search Summary

Data Type	Directory	Filename Pattern
Wheel Encoder Data	/beh	_encoder-data.csv
Psychopy Data	/beh	Psychopy CSV/JSON files
Meso Metadata	/func	_meso.ome.tiff_frame_metadata.json
Pupil Metadata	/func	_pupil.ome.tiff_frame_metadata.json
OME TIFF Stacks	/func	_meso.ome and _pupil.ome
Session-Level JSON files (for dropdown)	Parent directory above BIDS	*.json