%load_ext autoreload
%autoreload 2
import matplotlib.pyplot as plt
%matplotlib inline
import spatial_maps as sp
import septum_mec.analysis.data_processing as dp
import septum_mec.analysis.registration
import expipe
import os
import pathlib
import numpy as np
import exdir
import pandas as pd
import optogenetics as og
import quantities as pq
import shutil
from distutils.dir_util import copy_tree
import scipy
import scipy.signal as ss
from scipy.signal import find_peaks
from scipy.interpolate import interp1d
from matplotlib import mlab
from tqdm import tqdm_notebook as tqdm
from tqdm._tqdm_notebook import tqdm_notebook
tqdm_notebook.pandas()
data_loader = dp.Data()
actions = data_loader.actions
project = data_loader.project
#############################
perform_zscore = False
if not perform_zscore:
zscore_str = "-no-zscore"
else:
zscore_str = ""
#################################
output = pathlib.Path('output/stimulus-lfp-response' + zscore_str)
(output / 'data').mkdir(parents=True, exist_ok=True)
identify_neurons = actions['identify-neurons']
sessions = pd.read_csv(identify_neurons.data_path('sessions'))
channel_groups = []
for i, row in sessions.iterrows():
for ch in range(8):
row['channel_group'] = ch
channel_groups.append(row.to_dict())
channel_groups = pd.DataFrame(channel_groups)
def get_lim(action_id):
stim_times = data_loader.stim_times(action_id)
if stim_times is None:
return [0, np.inf]
stim_times = np.array(stim_times)
return [stim_times.min(), stim_times.max()]
def get_mask(lfp, lim):
return (lfp.times >= lim[0]) & (lfp.times <= lim[1])
def zscore(a):
return (a - a.mean()) / a.std()
def compute_stim_freq(action_id):
stim_times = data_loader.stim_times(action_id)
if stim_times is None:
return np.nan
stim_times = np.array(stim_times)
return 1 / np.mean(np.diff(stim_times))
def signaltonoise(a, axis=0, ddof=0):
a = np.asanyarray(a)
m = a.mean(axis)
sd = a.std(axis=axis, ddof=ddof)
return np.where(sd == 0, 0, m / sd)
def select_and_clean(anas, width=500, threshold=2):
anas = np.array(anas)
for ch in range(anas.shape[1]):
idxs, = np.where(abs(anas[:, ch]) > threshold)
for idx in idxs:
anas[idx-width:idx+width, ch] = 0 # TODO AR model prediction
return anas
def compute_energy(p, f, f1, f2):
if np.isnan(f1):
return np.nan
mask = (f > f1) & (f < f2)
df = f[1] - f[0]
return np.sum(p[mask]) * df
def compute_band_power(p, f, f1, f2):
if np.isnan(f1) or np.all(np.isnan(p)):
return [np.nan] * 2
from scipy.integrate import simps
dx = f[1] - f[0]
mask = (f > f1) & (f < f2)
# Compute the absolute power by approximating the area under the curve
band_power = simps(p[mask], dx=dx)
total_power = simps(p, dx=dx)
rel_power = band_power / total_power
return band_power, rel_power
def find_theta_peak(p, f, f1, f2):
if np.all(np.isnan(p)):
return np.nan, np.nan
mask = (f > f1) & (f < f2)
p_m = p[mask]
f_m = f[mask]
peaks, _ = find_peaks(p_m)
idx = np.argmax(p_m[peaks])
return f_m[peaks[idx]], p_m[peaks[idx]]
def compute_half_width(power, freq, max_power, max_frequency, band, band_width=1):
if np.isnan(max_power):
return [np.nan] * 3
# estimate baseline power
low_baseline_mask = (freq > band[0] - band_width) & (freq < band[0])
high_baseline_mask = (freq > band[1]) & (freq < band[1] + band_width)
baseline = np.mean(np.concatenate([power[low_baseline_mask], power[high_baseline_mask]]))
p = power - baseline
m_p = max_power - baseline
m_f = max_frequency
f = freq
# estimate half width
m_p_half = m_p / 2
half_p = p - m_p_half
idx_f = np.where(f <= m_f)[0].max()
idxs_p1, = np.where(np.diff(half_p[:idx_f + 1] > 0) == 1)
if len(idxs_p1) == 0:
return [np.nan] * 3
m1 = idxs_p1.max()
idxs_p2, = np.where(np.diff(half_p[idx_f:] > 0) == 1)
if len(idxs_p2) == 0:
return [np.nan] * 3
m2 = idxs_p2.min() + idx_f
# assert p[m1] < m_p_half < p[m1+1], (p[m1], m_p_half, p[m1+1])
# assert p[m2] > m_p_half > p[m2+1], (p[m2], m_p_half, p[m2+1])
f1 = interp1d([half_p[m1], half_p[m1 + 1]], [f[m1], f[m1 + 1]])(0)
f2 = interp1d([half_p[m2], half_p[m2 + 1]], [f[m2], f[m2 + 1]])(0)
return f1, f2, m_p_half + baseline
def compute_stim_peak(p, f, s_f):
if np.isnan(s_f):
return np.nan
return interp1d(f, p)(s_f)
def compute_relative_peak(power, freq, max_power, band, band_width=1):
# estimate baseline power
low_baseline_mask = (freq > band[0] - band_width) & (freq < band[0])
high_baseline_mask = (freq > band[1]) & (freq < band[1] + band_width)
baseline = np.mean(np.concatenate([power[low_baseline_mask], power[high_baseline_mask]]))
return (max_power - baseline) / abs(baseline)
theta_band_f1, theta_band_f2 = 6, 10
psd_data, freq_data = {}, {}
def process(row, perform_zscore):
action_id = row['action']
channel_group = row['channel_group']
name = f'{action_id}_{channel_group}'
lfp = data_loader.lfp(action_id, channel_group)
clean_lfp = select_and_clean(lfp)
snls = signaltonoise(clean_lfp)
best_channel = np.argmax(snls)
snl = snls[best_channel]
lim = get_lim(action_id)
mask = get_mask(lfp, lim)
if perform_zscore:
signal = zscore(clean_lfp[mask, best_channel].ravel())
else:
signal = clean_lfp[mask, best_channel].ravel()
window = int(6 * lfp.sampling_rate.magnitude)
# p_xx, freq = mlab.psd(signal, Fs=lfp.sampling_rate.magnitude, NFFT=NFFT)
freq, p_xx = ss.welch(signal, fs=lfp.sampling_rate.magnitude, nperseg=window, nfft=scipy.fftpack.next_fast_len(window))
# p_xx = 10 * np.log10(p_xx)
theta_f, theta_p_max = find_theta_peak(p_xx, freq, theta_band_f1, theta_band_f2)
theta_bandpower, theta_relpower = compute_band_power(p_xx, freq, theta_band_f1, theta_band_f2)
theta_relpeak = compute_relative_peak(p_xx, freq, theta_p_max, [theta_band_f1, theta_band_f2])
theta_half_f1, theta_half_f2, theta_half_power = compute_half_width(p_xx, freq, theta_p_max, theta_f, [theta_band_f1, theta_band_f2])
theta_half_width = theta_half_f2 - theta_half_f1
psd_data.update({name: p_xx})
freq_data.update({name: freq})
# stim
stim_freq = compute_stim_freq(action_id)
stim_p_max = compute_stim_peak(p_xx, freq, stim_freq)
stim_half_f1, stim_half_f2, stim_half_power = compute_half_width(p_xx, freq, stim_p_max, stim_freq, [stim_freq - 1, stim_freq + 1])
stim_half_width = stim_half_f2 - stim_half_f1
stim_bandpower, stim_relpower = compute_band_power(p_xx, freq, stim_freq - 1, stim_freq + 1)
stim_relpeak = compute_relative_peak(p_xx, freq, stim_p_max, [stim_freq - 1, stim_freq + 1])
result = pd.Series({
'signal_to_noise': snl,
'best_channel': best_channel,
'theta_freq': theta_f,
'theta_peak': theta_p_max,
'theta_bandpower': theta_bandpower,
'theta_relpower': theta_relpower,
'theta_relpeak': theta_relpeak,
'theta_half_f1': theta_half_f1,
'theta_half_f2': theta_half_f2,
'theta_half_width': theta_half_width,
'stim_freq': stim_freq,
'stim_p_max': stim_p_max,
'stim_half_f1': stim_half_f1,
'stim_half_f2': stim_half_f2,
'stim_half_width': stim_half_width,
'stim_bandpower': stim_bandpower,
'stim_relpower': stim_relpower,
'stim_relpeak': stim_relpeak,
})
return result
results = channel_groups.merge(
channel_groups.progress_apply(process, perform_zscore=perform_zscore, axis=1),
left_index=True, right_index=True)
pd.DataFrame(psd_data).to_feather(output / 'data' / 'psd.feather')
pd.DataFrame(freq_data).to_feather(output / 'data' / 'freqs.feather')
action = project.require_action("stimulus-lfp-response" + zscore_str)
action.modules['parameters'] = {
'window': 6,
'theta_band_f1': theta_band_f1,
'theta_band_f2': theta_band_f2
}
action.data['results'] = 'results.csv'
results.to_csv(action.data_path('results'), index=False)
copy_tree(output, str(action.data_path()))
septum_mec.analysis.registration.store_notebook(action, "10-calculate-stimulus-lfp-response.ipynb")