def run_analyses(spike_data_in, synced_spatial_data, opto_analysis=False):
snippet_data = PostSorting.load_snippet_data.get_snippets(
spike_data_in, prm, random_snippets=False)
spike_data = PostSorting.load_snippet_data.get_snippets(
spike_data_in, prm, random_snippets=True)
spike_data_spatial = PostSorting.open_field_spatial_firing.process_spatial_firing(
spike_data, synced_spatial_data)
spike_data_spatial = PostSorting.speed.calculate_speed_score(
synced_spatial_data, spike_data_spatial, 250, prm.get_sampling_rate())
hd_histogram, spatial_firing = PostSorting.open_field_head_direction.process_hd_data(
spike_data_spatial, synced_spatial_data, prm)
position_heat_map, spatial_firing = PostSorting.open_field_firing_maps.make_firing_field_maps(
synced_spatial_data, spike_data_spatial, prm)
spatial_firing = PostSorting.open_field_grid_cells.process_grid_data(
spatial_firing)
spatial_firing = PostSorting.open_field_firing_fields.analyze_firing_fields(
spatial_firing, synced_spatial_data, prm)
if opto_analysis:
PostSorting.open_field_light_data.process_spikes_around_light(
spike_data_spatial, prm)
save_data_frames(spatial_firing,
synced_spatial_data,
snippet_data=snippet_data)
save_data_for_plots(position_heat_map, hd_histogram, prm)
make_plots(synced_spatial_data, spatial_firing, position_heat_map,
hd_histogram, prm)
return synced_spatial_data, spatial_firing
def process_data(analysis_path):
organize_data(analysis_path)
for name in glob.glob(analysis_path + '*'):
if os.path.isdir(name):
if os.path.exists(name + '/spatial_firing.pkl'):
print(name)
prm.set_file_path(name)
prm.set_output_path(name)
position = pd.read_pickle(name + '/position.pkl')
# process position data - add hd etc
spatial_firing = pd.read_pickle(name + '/spatial_firing.pkl')
hd = [item for sublist in spatial_firing.hd[0] for item in sublist]
spatial_firing['hd'] = [np.array(hd) - 180]
#if len(spatial_firing.hd) == 1:
# spatial_firing['hd'] = np.array(spatial_firing.hd)
spatial_firing['position_x_pixels'] = spatial_firing.position_x
spatial_firing['position_y_pixels'] = spatial_firing.position_y
prm.set_sampling_rate(1000000) # this is to make the histograms similar to the real data
hd_histogram, spatial_firing = PostSorting.open_field_head_direction.process_hd_data(spatial_firing, position, prm)
# if 'firing_maps' not in spatial_firing:
position_heat_map, spatial_firing = get_rate_maps(position, spatial_firing)
spatial_firing = PostSorting.open_field_grid_cells.process_grid_data(spatial_firing)
spatial_firing = PostSorting.open_field_firing_fields.analyze_firing_fields(spatial_firing, position, prm)
spatial_firing.to_pickle(name + '/spatial_firing.pkl')
make_plots(position, spatial_firing, position_heat_map, hd_histogram, prm)
def post_process_recording(recording_to_process,
session_type,
sorter_name='MountainSort'):
create_folders_for_output(recording_to_process)
initialize_parameters(recording_to_process)
prm.set_sorter_name('/' + sorter_name)
prm.set_output_path(recording_to_process + prm.get_sorter_name())
#Process position
print('process position')
raw_position_data, processed_position_data = process_position_data(
recording_to_process, prm)
print('process file properties')
#save data
if os.path.exists(prm.get_output_path() + '/DataFrames') is False:
os.makedirs(prm.get_output_path() + '/DataFrames')
# raw_position_data.to_pickle(prm.get_output_path() + '/DataFrames/raw_position_data.pkl')
# processed_position_data.to_pickle(prm.get_output_path() + '/DataFrames/processed_position_data.pkl')
# processed_position_data = pd.read_pickle(prm.get_output_path() + '/DataFrames/processed_position_data.pkl')
#Process firing
spike_data, bad_clusters = process_firing_properties(
recording_to_process, session_type, prm)
if len(
spike_data
) == 0: # this means that there are no good clusters and the analysis will not run
save_data_frames(prm, spike_data, raw_position_data,
processed_position_data, bad_clusters)
return
spike_data = PostSorting.load_snippet_data.get_snippets(spike_data, prm)
spike_data = PostSorting.vr_spatial_firing.process_spatial_firing(
spike_data, raw_position_data)
#spike_data = PostSorting.vr_firing_rate_maps.make_firing_field_maps_for_trial_types(spike_data, raw_position_data, processed_position_data)
spike_data = PostSorting.vr_firing_rate_maps.make_firing_field_maps_all(
spike_data, raw_position_data, processed_position_data)
spike_data = PostSorting.vr_FiringMaps_InTime.control_convolution_in_time(
spike_data, raw_position_data)
#spike_data = PostSorting.vr_firing_maps_copy.make_firing_field_maps(raw_position_data, spike_data, prm)
save_data_frames(prm, spike_data, raw_position_data,
processed_position_data, bad_clusters)
make_plots(spike_data, raw_position_data, processed_position_data)
gc.collect()
def call_stable_functions(recording_to_process, session_type, analysis_type):
# process opto data -this has to be done before splitting the session into recording and opto-tagging parts
opto_on, opto_off, opto_is_found = process_light_stimulation(
recording_to_process, prm)
# process spatial data
spatial_data, position_was_found = process_position_data(
recording_to_process, session_type, prm)
if position_was_found:
synced_spatial_data = sync_data(recording_to_process, prm,
spatial_data)
# analyze spike data
spike_data = PostSorting.load_firing_data.create_firing_data_frame(
recording_to_process, session_type, prm)
spike_data = PostSorting.temporal_firing.add_temporal_firing_properties_to_df(
spike_data, prm)
if analysis_type is 'default':
spike_data, bad_clusters = PostSorting.curation.curate_data(
spike_data, prm)
if len(
spike_data
) == 0: # this means that there are no good clusters and the analysis will not run
save_data_frames(spike_data,
synced_spatial_data,
bad_clusters=bad_clusters)
return
spike_data = PostSorting.load_snippet_data.get_snippets(
spike_data, prm)
spike_data_spatial = PostSorting.open_field_spatial_firing.process_spatial_firing(
spike_data, synced_spatial_data)
if opto_is_found:
PostSorting.open_field_light_data.process_spikes_around_light(
spike_data_spatial, prm)
spike_data_spatial = PostSorting.speed.calculate_speed_score(
synced_spatial_data, spike_data_spatial, 250,
prm.get_sampling_rate())
hd_histogram, spatial_firing = PostSorting.open_field_head_direction.process_hd_data(
spike_data_spatial, synced_spatial_data, prm)
position_heat_map, spatial_firing = PostSorting.open_field_firing_maps.make_firing_field_maps(
synced_spatial_data, spike_data_spatial, prm)
spatial_firing = PostSorting.open_field_grid_cells.process_grid_data(
spatial_firing)
spatial_firing = PostSorting.open_field_firing_fields.analyze_firing_fields(
spatial_firing, synced_spatial_data, prm)
save_data_frames(spatial_firing, synced_spatial_data)
make_plots(synced_spatial_data, spatial_firing, position_heat_map,
hd_histogram, prm)
def run_analyses(spike_data_in, synced_spatial_data):
spike_data = PostSorting.load_snippet_data.get_snippets(spike_data_in, prm)
spike_data_spatial = PostSorting.open_field_spatial_firing.process_spatial_firing(
spike_data, synced_spatial_data)
spike_data_spatial = PostSorting.speed.calculate_speed_score(
synced_spatial_data, spike_data_spatial, 250, prm.get_sampling_rate())
hd_histogram, spatial_firing = PostSorting.open_field_head_direction.process_hd_data(
spike_data_spatial, synced_spatial_data, prm)
position_heat_map, spatial_firing = PostSorting.open_field_firing_maps.make_firing_field_maps(
synced_spatial_data, spike_data_spatial, prm)
spatial_firing = PostSorting.open_field_grid_cells.process_grid_data(
spatial_firing)
spatial_firing = PostSorting.open_field_firing_fields.analyze_firing_fields(
spatial_firing, synced_spatial_data, prm)
save_data_frames(spatial_firing, synced_spatial_data)
make_plots(synced_spatial_data, spatial_firing, position_heat_map,
hd_histogram, prm)
return synced_spatial_data, spatial_firing
def process_data(folder_to_search_in):
prm.set_sampling_rate(
48000) # this is according to Sargolini et al. (2006)
prm.set_pixel_ratio(
100
) # this is because the data is already in cm so there's no need to convert
prm.set_sorter_name('Manual')
# prm.set_is_stable(True) # todo: this needs to be removed - R analysis won't run for now
for name in glob.glob(folder_to_search_in + '/*.mat'):
if os.path.exists(name):
if 'POS' in name:
print('I found this:' + name)
position_data_matlab = loadmat(name)
position_data = get_position_data_frame(position_data_matlab)
session_id = name.split('\\')[-1].split('_')[0]
if position_data is not False:
try:
firing_data = pd.DataFrame()
create_folder_structure(name, session_id, prm)
firing_data = fill_firing_data_frame(
position_data, firing_data, name,
folder_to_search_in, session_id)
hd_histogram, spatial_firing = PostSorting.open_field_head_direction.process_hd_data(
firing_data, position_data, prm)
position_heat_map, spatial_firing = get_rate_maps(
position_data, firing_data)
spatial_firing = PostSorting.open_field_grid_cells.process_grid_data(
spatial_firing)
spatial_firing = PostSorting.open_field_firing_fields.analyze_firing_fields(
spatial_firing, position_data, prm)
save_data_frames(spatial_firing, position_data)
make_plots(position_data, spatial_firing,
position_heat_map, hd_histogram, prm)
except Exception as ex:
print('I failed to analyze this cell.')
print(ex)
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_tb(exc_traceback)
print('Processing finished.')
def main():
print('-------------------------------------------------------------')
print('-------------------------------------------------------------')
prm = PostSorting.parameters.Parameters()
prm.set_pixel_ratio(440)
prm.set_sampling_rate(30000)
path_to_recordings = '/M5_2018-03-06_15-34-44_of'
path_to_recordings = '/recordings/test_figures'
prm.set_output_path(path_to_recordings + '/MountainSort/')
position_data = pd.read_pickle(path_to_recordings +
'/MountainSort/DataFrames/position.pkl')
spatial_firing = pd.read_pickle(
path_to_recordings + '/MountainSort/DataFrames/spatial_firing.pkl')
position_heat_map = np.load(
path_to_recordings + '/MountainSort/DataFrames/position_heat_map.npy')
hd_histogram = np.load(path_to_recordings +
'/MountainSort/DataFrames/hd_histogram.npy')
make_plots(position_data, spatial_firing, position_heat_map, hd_histogram,
prm)