def test_psth_between_sweep_interpolation():
feature = "test_feature_name"
test_spike_times = ([0.25, 0.75], [0.2, 0.5, 0.6, 0.7])
available_list = []
for tst in test_spike_times:
spike_info = pd.DataFrame({
"threshold_t": tst,
})
available_list.append(spike_info)
start = 0
end = 1
width = 20
n_bins = int((end - start) / (width * 0.001))
si_list = [None, available_list[0], None, available_list[1], None]
output = fv.psth_vector(si_list, start=start, end=end, width=width)
assert np.array_equal(output[:n_bins], output[n_bins:2 * n_bins])
assert np.array_equal(output[3 * n_bins:4 * n_bins],
output[4 * n_bins:5 * n_bins])
assert np.array_equal(
output[2 * n_bins:3 * n_bins],
np.vstack([output[n_bins:2 * n_bins],
output[4 * n_bins:5 * n_bins]]).mean(axis=0))
def test_psth_compressed_firing():
test_spike_times = [0.2, 0.202, 0.204]
spike_info = pd.DataFrame({"threshold_t": test_spike_times})
start = 0
end = 1
width = 50
# All spikes are within one bin
output = fv.psth_vector([spike_info], start=start, end=end, width=width)
assert np.sum(output > 0) == 1
assert np.max(output) == len(test_spike_times) / (width * 0.001)
def test_psth_number_of_spikes():
np.random.seed(42)
n_spikes = np.random.randint(0, 100)
start = 1.02
end = 2.02
width = 50
test_spike_times = np.random.random(n_spikes) * (end - start) + start
spike_info = pd.DataFrame({"threshold_t": test_spike_times})
output = fv.psth_vector([spike_info], start=start, end=end, width=width)
assert np.isclose(output.mean(), n_spikes)
def test_psth_sparse_firing():
test_spike_times = [0.2, 0.5]
spike_info = pd.DataFrame({"threshold_t": test_spike_times})
start = 0
end = 1
width = 50
# All spikes are in own bins
output = fv.psth_vector([spike_info], start=start, end=end, width=width)
assert np.sum(output > 0) == len(test_spike_times)
assert np.max(output) == 1 / (width * 0.001)
def test_psth_duration_rounding():
start_a, end_a = 1.02, 2.02
start_b, end_b = 1.02, 2.0199999999999996
np.random.seed(42)
n_spikes = np.random.randint(0, 100)
test_spike_times = np.random.random(n_spikes) * (end_a - start_a) + start_a
spike_info = pd.DataFrame({"threshold_t": test_spike_times})
width = 50
output_a = fv.psth_vector([spike_info],
start=start_a,
end=end_a,
width=width)
output_b = fv.psth_vector([spike_info],
start=start_b,
end=end_b,
width=width)
assert output_a.shape == output_b.shape
def data_for_specimen_id(specimen_id,
sweep_qc_option,
data_source,
ap_window_length=0.006,
target_sampling_rate=10000,
nfiles=None):
logging.debug("specimen_id: {}".format(specimen_id))
lsq_fail = False
ssq_fail = False
ramp_fail = False
# Find or retrieve NWB file and ancillary info and construct an AibsDataSet object
ontology = StimulusOntology(
ju.read(StimulusOntology.DEFAULT_STIMULUS_ONTOLOGY_FILE))
if data_source == "local":
nwb_path = nfiles[specimen_id]
if type(nwb_path) is dict and "error" in nwb_path:
logging.warning(
"Problem getting NWB file for specimen {:d}".format(
specimen_id))
return nwb_path
data_set = HBGDataSet(nwb_file=nwb_path, ontology=ontology)
else:
logging.error("invalid data source specified ({})".format(data_source))
# Identify and preprocess long square sweeps
try:
lsq_sweep_numbers = categorize_iclamp_sweeps(
data_set,
ontology.long_square_names,
sweep_qc_option=sweep_qc_option,
specimen_id=specimen_id)
(lsq_sweeps, lsq_features, lsq_start, lsq_end,
lsq_spx) = preprocess_long_square_sweeps(data_set, lsq_sweep_numbers)
except Exception as detail:
lsq_fail = True
logging.warning(
"Exception when preprocessing long square sweeps from specimen {:d}"
.format(specimen_id))
logging.warning(detail)
return {
"error": {
"type": "sweep_table",
"details": traceback.format_exc(limit=None)
}
}
# Identify and preprocess short square sweeps
try:
ssq_sweep_numbers = categorize_iclamp_sweeps(
data_set,
ontology.short_square_names,
sweep_qc_option=sweep_qc_option,
specimen_id=specimen_id)
ssq_sweeps, ssq_features = preprocess_short_square_sweeps(
data_set, ssq_sweep_numbers)
except Exception as detail:
ssq_fail = True
logging.warning(
"Exception when preprocessing short square sweeps from specimen {:d}"
.format(specimen_id))
logging.warning(detail)
{
"error": {
"type": "sweep_table",
"details": traceback.format_exc(limit=None)
}
}
# Identify and preprocess ramp sweeps
try:
ramp_sweep_numbers = categorize_iclamp_sweeps(
data_set,
ontology.ramp_names,
sweep_qc_option=sweep_qc_option,
specimen_id=specimen_id)
ramp_sweeps, ramp_features = preprocess_ramp_sweeps(
data_set, ramp_sweep_numbers)
except Exception as detail:
ramp_fail = True
logging.warning(
"Exception when preprocessing ramp sweeps from specimen {:d}".
format(specimen_id))
logging.warning(detail)
{
"error": {
"type": "sweep_table",
"details": traceback.format_exc(limit=None)
}
}
# Calculate desired feature vectors
result = {}
try:
(subthresh_hyperpol_dict, hyperpol_deflect_dict
) = fv.identify_subthreshold_hyperpol_with_amplitudes(
lsq_features, lsq_sweeps)
target_amps_for_step_subthresh = [-90, -70, -50, -30, -10]
result["step_subthresh"] = fv.step_subthreshold(
subthresh_hyperpol_dict,
target_amps_for_step_subthresh,
lsq_start,
lsq_end,
amp_tolerance=5)
result["subthresh_norm"] = fv.subthresh_norm(subthresh_hyperpol_dict,
hyperpol_deflect_dict,
lsq_start, lsq_end)
(subthresh_depol_dict,
depol_deflect_dict) = fv.identify_subthreshold_depol_with_amplitudes(
lsq_features, lsq_sweeps)
result["subthresh_depol_norm"] = fv.subthresh_depol_norm(
subthresh_depol_dict, depol_deflect_dict, lsq_start, lsq_end)
isi_sweep, isi_sweep_spike_info = fv.identify_sweep_for_isi_shape(
lsq_sweeps, lsq_features, lsq_end - lsq_start)
result["isi_shape"] = fv.isi_shape(isi_sweep, isi_sweep_spike_info,
lsq_end)
if ssq_fail == False:
# Calculate waveforms from each type of sweep
spiking_ssq_sweep_list = [
ssq_sweeps.sweeps[swp_ind]
for swp_ind in ssq_features["common_amp_sweeps"].index
]
spiking_ssq_info_list = [
ssq_features["spikes_set"][swp_ind]
for swp_ind in ssq_features["common_amp_sweeps"].index
]
ssq_ap_v, ssq_ap_dv = fv.first_ap_vectors(
spiking_ssq_sweep_list,
spiking_ssq_info_list,
target_sampling_rate=target_sampling_rate,
window_length=ap_window_length,
skip_clipped=True)
else:
ssq_ap_v, ssq_ap_dv = np.nan, np.nan
rheo_ind = lsq_features["rheobase_sweep"].name
sweep = lsq_sweeps.sweeps[rheo_ind]
lsq_ap_v, lsq_ap_dv = fv.first_ap_vectors(
[sweep], [lsq_features["spikes_set"][rheo_ind]],
target_sampling_rate=target_sampling_rate,
window_length=ap_window_length)
if ramp_fail == False:
spiking_ramp_sweep_list = [
ramp_sweeps.sweeps[swp_ind]
for swp_ind in ramp_features["spiking_sweeps"].index
]
spiking_ramp_info_list = [
ramp_features["spikes_set"][swp_ind]
for swp_ind in ramp_features["spiking_sweeps"].index
]
ramp_ap_v, ramp_ap_dv = fv.first_ap_vectors(
spiking_ramp_sweep_list,
spiking_ramp_info_list,
target_sampling_rate=target_sampling_rate,
window_length=ap_window_length,
skip_clipped=True)
else:
ramp_ap_v, ramp_ap_dv = np.nan, np.nan
if ramp_fail == True:
ramp_ap_dv = np.copy(lsq_ap_dv)
ramp_ap_v = np.copy(lsq_ap_v)
if ssq_fail == True:
ssq_ap_dv = np.copy(lsq_ap_dv)
ssq_ap_v = np.copy(lsq_ap_v)
# Combine so that differences can be assessed by analyses like sPCA
result["first_ap_v"] = np.hstack([ssq_ap_v, lsq_ap_v, ramp_ap_v])
result["first_ap_dv"] = np.hstack([ssq_ap_dv, lsq_ap_dv, ramp_ap_dv])
target_amplitudes = np.arange(0, 120, 20)
supra_info_list = fv.identify_suprathreshold_spike_info(
lsq_features, target_amplitudes, shift=10)
result["psth"] = fv.psth_vector(supra_info_list, lsq_start, lsq_end)
result["inst_freq"] = fv.inst_freq_vector(supra_info_list, lsq_start,
lsq_end)
spike_feature_list = [
"upstroke_downstroke_ratio",
"peak_v",
"fast_trough_v",
"threshold_v",
"width",
]
for feature in spike_feature_list:
result["spiking_" + feature] = fv.spike_feature_vector(
feature, supra_info_list, lsq_start, lsq_end)
if feature == 'width':
result["spiking_width"] = result["spiking_width"] / 2
except Exception as detail:
logging.warning(
"Exception when processing specimen {:d}".format(specimen_id))
logging.warning(detail)
return {
"error": {
"type": "processing",
"details": traceback.format_exc(limit=None)
}
}
return result
def main(nwb_file, output_dir, project, **kwargs):
nwb = MiesNwb(nwb_file)
# SPECIFICS FOR EXAMPLE NWB =========
# Only analyze one channel at a time
channel = 0
# We can work out code to automatically extract these based on stimulus names later.
if_sweep_inds = [39, 45]
targetv_sweep_inds = [15, 21]
# END SPECIFICS =====================
# Assemble all Recordings and convert to Sweeps
supra_sweep_ids = list(range(*if_sweep_inds))
sub_sweep_ids = list(range(*targetv_sweep_inds))
supra_recs = [nwb.contents[i][channel] for i in supra_sweep_ids]
sub_recs = [nwb.contents[i][channel] for i in sub_sweep_ids]
# Build sweep sets
lsq_supra_sweep_list, lsq_supra_dur = recs_to_sweeps(supra_recs)
lsq_sub_sweep_list, lsq_sub_dur = recs_to_sweeps(sub_recs)
lsq_supra_sweeps = SweepSet(lsq_supra_sweep_list)
lsq_sub_sweeps = SweepSet(lsq_sub_sweep_list)
lsq_supra_start = 0
lsq_supra_end = lsq_supra_dur
lsq_sub_start = 0
lsq_sub_end = lsq_sub_dur
# Pre-process sweeps
lsq_supra_spx, lsq_supra_spfx = dsf.extractors_for_sweeps(
lsq_supra_sweeps, start=lsq_supra_start, end=lsq_supra_end)
lsq_supra_an = spa.LongSquareAnalysis(lsq_supra_spx,
lsq_supra_spfx,
subthresh_min_amp=-100.,
require_subthreshold=False)
lsq_supra_features = lsq_supra_an.analyze(lsq_supra_sweeps)
lsq_sub_spx, lsq_sub_spfx = dsf.extractors_for_sweeps(lsq_sub_sweeps,
start=lsq_sub_start,
end=lsq_sub_end)
lsq_sub_an = spa.LongSquareAnalysis(lsq_sub_spx,
lsq_sub_spfx,
subthresh_min_amp=-100.,
require_suprathreshold=False)
lsq_sub_features = lsq_sub_an.analyze(lsq_sub_sweeps)
# Calculate feature vectors
result = {}
(subthresh_hyperpol_dict, hyperpol_deflect_dict
) = fv.identify_subthreshold_hyperpol_with_amplitudes(
lsq_sub_features, lsq_sub_sweeps)
target_amps_for_step_subthresh = [-90, -70, -50, -30, -10]
result["step_subthresh"] = fv.step_subthreshold(
subthresh_hyperpol_dict,
target_amps_for_step_subthresh,
lsq_sub_start,
lsq_sub_end,
amp_tolerance=5)
result["subthresh_norm"] = fv.subthresh_norm(subthresh_hyperpol_dict,
hyperpol_deflect_dict,
lsq_sub_start, lsq_sub_end)
(subthresh_depol_dict,
depol_deflect_dict) = fv.identify_subthreshold_depol_with_amplitudes(
lsq_supra_features, lsq_supra_sweeps)
result["subthresh_depol_norm"] = fv.subthresh_depol_norm(
subthresh_depol_dict, depol_deflect_dict, lsq_supra_start,
lsq_supra_end)
isi_sweep, isi_sweep_spike_info = fv.identify_sweep_for_isi_shape(
lsq_supra_sweeps, lsq_supra_features, lsq_supra_end - lsq_supra_start)
result["isi_shape"] = fv.isi_shape(isi_sweep, isi_sweep_spike_info,
lsq_supra_end)
# Calculate AP waveform from long squares
rheo_ind = lsq_supra_features["rheobase_sweep"].name
sweep = lsq_supra_sweeps.sweeps[rheo_ind]
lsq_ap_v, lsq_ap_dv = fv.first_ap_vectors(
[sweep], [lsq_supra_features["spikes_set"][rheo_ind]],
window_length=ap_window_length)
result["first_ap_v"] = lsq_ap_v
result["first_ap_dv"] = lsq_ap_dv
target_amplitudes = np.arange(0, 120, 20)
supra_info_list = fv.identify_suprathreshold_sweep_sequence(
lsq_supra_features, target_amplitudes, shift=10)
result["psth"] = fv.psth_vector(supra_info_list, lsq_supra_start,
lsq_supra_end)
result["inst_freq"] = fv.inst_freq_vector(supra_info_list, lsq_supra_start,
lsq_supra_end)
spike_feature_list = [
"upstroke_downstroke_ratio",
"peak_v",
"fast_trough_v",
"threshold_v",
"width",
]
for feature in spike_feature_list:
result["spiking_" + feature] = fv.spike_feature_vector(
feature, supra_info_list, lsq_supra_start, lsq_supra_end)
# Save the results
specimen_ids = [0]
results = [result]
filtered_set = [(i, r) for i, r in zip(specimen_ids, results)
if not "error" in r.keys()]
error_set = [{
"id": i,
"error": d
} for i, d in zip(specimen_ids, results) if "error" in d.keys()]
if len(filtered_set) == 0:
logging.info("No specimens had results")
return
with open(os.path.join(output_dir, "fv_errors_{:s}.json".format(project)),
"w") as f:
json.dump(error_set, f, indent=4)
used_ids, results = zip(*filtered_set)
logging.info("Finished with {:d} processed specimens".format(
len(used_ids)))
k_sizes = {}
for k in results[0].keys():
if k not in k_sizes and results[0][k] is not None:
k_sizes[k] = len(results[0][k])
data = np.array([
r[k] if k in r else np.nan * np.zeros(k_sizes[k]) for r in results
])
if len(data.shape) == 1: # it'll be 1D if there's just one specimen
data = np.reshape(data, (1, -1))
if data.shape[0] < len(used_ids):
logging.warn("Missing data!")
missing = np.array([k not in r for r in results])
print(k, np.array(used_ids)[missing])
np.save(
os.path.join(output_dir, "fv_{:s}_{:s}.npy".format(k, project)),
data)
np.save(os.path.join(output_dir, "fv_ids_{:s}.npy".format(project)),
used_ids)
def data_for_specimen_id(
specimen_id,
sweep_qc_option,
data_source,
ontology,
ap_window_length=0.005,
target_sampling_rate=50000,
file_list=None,
):
"""
Extract feature vector from given cell identified by the specimen_id
Parameters
----------
specimen_id : int
cell identified
sweep_qc_option : str
see CollectFeatureVectorParameters input schema for details
data_source: str
see CollectFeatureVectorParameters input schema for details
ontology : stimulus.StimulusOntology
mapping of stimuli names to stimulus codes
ap_window_length : float
see CollectFeatureVectorParameters input schema for details
target_sampling_rate : float
sampling rate
file_list : list of str
nwbfile names
Returns
-------
dict :
features for a given cell specimen_id
"""
logging.debug("specimen_id: {}".format(specimen_id))
# Find or retrieve NWB file and ancillary info and construct an AibsDataSet object
data_set = su.dataset_for_specimen_id(specimen_id, data_source, ontology,
file_list)
if type(data_set) is dict and "error" in data_set:
logging.warning(
"Problem getting AibsDataSet for specimen {:d} from LIMS".format(
specimen_id))
return data_set
# Identify and preprocess long square sweeps
try:
lsq_sweep_numbers = su.categorize_iclamp_sweeps(
data_set,
ontology.long_square_names,
sweep_qc_option=sweep_qc_option,
specimen_id=specimen_id)
(lsq_sweeps, lsq_features, _, lsq_start,
lsq_end) = su.preprocess_long_square_sweeps(data_set,
lsq_sweep_numbers)
except Exception as detail:
logging.warning(
"Exception when preprocessing long square sweeps from specimen {:d}"
.format(specimen_id))
logging.warning(detail)
return {
"error": {
"type": "sweep_table",
"details": traceback.format_exc(limit=None)
}
}
# Identify and preprocess short square sweeps
try:
ssq_sweep_numbers = su.categorize_iclamp_sweeps(
data_set,
ontology.short_square_names,
sweep_qc_option=sweep_qc_option,
specimen_id=specimen_id)
ssq_sweeps, ssq_features, _ = su.preprocess_short_square_sweeps(
data_set, ssq_sweep_numbers)
except Exception as detail:
logging.warning(
"Exception when preprocessing short square sweeps from specimen {:d}"
.format(specimen_id))
logging.warning(detail)
return {
"error": {
"type": "sweep_table",
"details": traceback.format_exc(limit=None)
}
}
# Identify and preprocess ramp sweeps
try:
ramp_sweep_numbers = su.categorize_iclamp_sweeps(
data_set,
ontology.ramp_names,
sweep_qc_option=sweep_qc_option,
specimen_id=specimen_id)
ramp_sweeps, ramp_features, _ = su.preprocess_ramp_sweeps(
data_set, ramp_sweep_numbers)
except Exception as detail:
logging.warning(
"Exception when preprocessing ramp sweeps from specimen {:d}".
format(specimen_id))
logging.warning(detail)
return {
"error": {
"type": "sweep_table",
"details": traceback.format_exc(limit=None)
}
}
# Calculate desired feature vectors
result = {}
if data_source == "filesystem":
result["id"] = [specimen_id]
try:
(subthresh_hyperpol_dict, hyperpol_deflect_dict
) = fv.identify_subthreshold_hyperpol_with_amplitudes(
lsq_features, lsq_sweeps)
target_amps_for_step_subthresh = [-90, -70, -50, -30, -10]
result["step_subthresh"] = fv.step_subthreshold(
subthresh_hyperpol_dict,
target_amps_for_step_subthresh,
lsq_start,
lsq_end,
amp_tolerance=5)
result["subthresh_norm"] = fv.subthresh_norm(subthresh_hyperpol_dict,
hyperpol_deflect_dict,
lsq_start, lsq_end)
(subthresh_depol_dict,
depol_deflect_dict) = fv.identify_subthreshold_depol_with_amplitudes(
lsq_features, lsq_sweeps)
result["subthresh_depol_norm"] = fv.subthresh_depol_norm(
subthresh_depol_dict, depol_deflect_dict,
np.round(lsq_start, decimals=3), np.round(lsq_end, decimals=3))
isi_sweep, isi_sweep_spike_info = fv.identify_sweep_for_isi_shape(
lsq_sweeps, lsq_features, lsq_end - lsq_start)
result["isi_shape"] = fv.isi_shape(isi_sweep, isi_sweep_spike_info,
lsq_end)
# Calculate waveforms from each type of sweep
spiking_ssq_sweep_list = [
ssq_sweeps.sweeps[swp_ind]
for swp_ind in ssq_features["common_amp_sweeps"].index
]
spiking_ssq_info_list = [
ssq_features["spikes_set"][swp_ind]
for swp_ind in ssq_features["common_amp_sweeps"].index
]
ssq_ap_v, ssq_ap_dv = fv.first_ap_vectors(
spiking_ssq_sweep_list,
spiking_ssq_info_list,
target_sampling_rate=target_sampling_rate,
window_length=ap_window_length,
skip_clipped=True)
rheo_ind = lsq_features["rheobase_sweep"].name
sweep = lsq_sweeps.sweeps[rheo_ind]
lsq_ap_v, lsq_ap_dv = fv.first_ap_vectors(
[sweep], [lsq_features["spikes_set"][rheo_ind]],
target_sampling_rate=target_sampling_rate,
window_length=ap_window_length)
spiking_ramp_sweep_list = [
ramp_sweeps.sweeps[swp_ind]
for swp_ind in ramp_features["spiking_sweeps"].index
]
spiking_ramp_info_list = [
ramp_features["spikes_set"][swp_ind]
for swp_ind in ramp_features["spiking_sweeps"].index
]
ramp_ap_v, ramp_ap_dv = fv.first_ap_vectors(
spiking_ramp_sweep_list,
spiking_ramp_info_list,
target_sampling_rate=target_sampling_rate,
window_length=ap_window_length,
skip_clipped=True)
# Combine so that differences can be assessed by analyses like sPCA
result["first_ap_v"] = np.hstack([ssq_ap_v, lsq_ap_v, ramp_ap_v])
result["first_ap_dv"] = np.hstack([ssq_ap_dv, lsq_ap_dv, ramp_ap_dv])
target_amplitudes = np.arange(0, 120, 20)
supra_info_list = fv.identify_suprathreshold_spike_info(
lsq_features, target_amplitudes, shift=10)
result["psth"] = fv.psth_vector(supra_info_list, lsq_start, lsq_end)
result["inst_freq"] = fv.inst_freq_vector(supra_info_list, lsq_start,
lsq_end)
spike_feature_list = [
"upstroke_downstroke_ratio",
"peak_v",
"fast_trough_v",
"threshold_v",
"width",
]
for feature in spike_feature_list:
result["spiking_" + feature] = fv.spike_feature_vector(
feature, supra_info_list, lsq_start, lsq_end)
except Exception as detail:
logging.warning(
"Exception when processing specimen {:d}".format(specimen_id))
logging.warning(detail)
return {
"error": {
"type": "processing",
"details": traceback.format_exc(limit=None)
}
}
return result
