utils.py

from glob import glob
import pandas as pd
import os
from pathlib import Path
import numpy as np
import pickle

def list_sessions(animal,data_path="/ceph/imaging1/davide/2p_data"):
    session_list = glob(data_path + f'/{animal}/*/', recursive = True)
    
    return session_list

def list_complete_sessions(animal,data_path="/ceph/imaging1/davide/2p_data"):
    session_list = glob(data_path + f'/{animal}/*/', recursive = True)
    complete_sessions = [s for s in session_list if check_complete_session(s)]
    return complete_sessions

def check_neural_data_file(folder_path):
    file_path = os.path.join(folder_path, 'neural_data.pickle')
    return os.path.isfile(file_path)

def check_vr_data_file(folder_path):
    file_path = os.path.join(folder_path, 'behaviour_data.pickle')
    return os.path.isfile(file_path)

def check_trial_data_file(folder_path):
    file_path = os.path.join(folder_path, 'trial_data.csv')
    return os.path.isfile(file_path)

def check_complete_session(folder_path):
    return check_neural_data_file(folder_path) and check_trial_data_file(folder_path) and check_vr_data_file(folder_path)


def print_animal_summary(animal,data_path="/ceph/imaging1/davide/2p_data"):
    '''
    Gives a summary of the preprocessed data for the given animal
    '''
    
    scanner_fps = 30.
    behaviour_fps = 1000.
    
    print(f'Overview for {animal}')
    
    sessions = list_sessions(animal)
    complete_sessions = list_complete_sessions(animal)
    print(f'# of sessions: {len(sessions)}, of which {len(complete_sessions)} complete')
    print('\n')
    
    for s in sessions:
        print(f"date: {s.split('/')[-2]}")
        
        ## NEURAL DATA
        
        try:
            with open(Path(s).joinpath(f'neural_data.pickle'),'rb') as pfile:
                n_data = pickle.load(pfile)
                n_cells = n_data['traces'].shape[0]
                recording_duration =  n_data['traces'].shape[1]/scanner_fps

        except FileNotFoundError:
            n_cells = np.nan
            recording_duration = np.nan
            
        
        ## BEHAVIOUR DATA
        
        try:
            trial_data = pd.read_csv(Path(s).joinpath('trial_data.csv'))
            n_trials = len(trial_data)

        except FileNotFoundError:
            n_trials = np.nan
            
            
        try:
            with open(Path(s).joinpath(f'behaviour_data.pickle'),'rb') as pfile:
                    b_data = pickle.load(pfile)
                    vr_duration = b_data['time'].shape[0]/behaviour_fps
                    
        except FileNotFoundError:
            vr_duration = np.nan
            

        print(f' NEURAL DATA -cells: {n_cells}, rec duration: {recording_duration:.2f} s')
        print(f' VR DATA -trials: {n_trials}, vr duration: {vr_duration:.2f} s')
        print('\n')