from CellData import CellData
import numpy as np
from scipy.optimize import curve_fit
from scipy.stats import linregress
import matplotlib.pyplot as plt
from warnings import warn
import functions as fu
class FICurve:
def __init__(self, cell_data: CellData, contrast: bool = True):
self.cell_data = cell_data
self.using_contrast = contrast
if contrast:
self.stimulus_value = cell_data.get_fi_contrasts()
else:
self.stimulus_value = cell_data.get_fi_intensities()
self.f_zeros = []
self.f_infinities = []
self.f_baselines = []
# f_max, f_min, k, x_zero
self.boltzmann_fit_vars = []
# increase, offset
self.f_infinity_fit = []
self.all_calculate_frequency_points()
self.fit_line()
self.fit_boltzmann()
def all_calculate_frequency_points(self):
mean_frequencies = self.cell_data.get_mean_isi_frequencies()
time_axes = self.cell_data.get_time_axes_mean_frequencies()
if len(mean_frequencies) == 0:
warn("FICurve:all_calculate_frequency_points(): mean_frequencies is empty.\n"
"Was all_calculate_mean_isi_frequencies already called?")
for i in range(len(mean_frequencies)):
if time_axes[i][0] > self.cell_data.get_stimulus_start():
warn("TODO: Deal with to strongly cut frequency traces in cell data! ")
self.f_zeros.append(-1)
self.f_baselines.append(-1)
self.f_infinities.append(-1)
continue
self.f_zeros.append(self.__calculate_f_zero__(time_axes[i], mean_frequencies[i]))
self.f_baselines.append(self.__calculate_f_baseline__(time_axes[i], mean_frequencies[i]))
self.f_infinities.append(self.__calculate_f_infinity__(time_axes[i], mean_frequencies[i]))
def fit_line(self):
popt, pcov = curve_fit(fu.clipped_line, self.stimulus_value, self.f_infinities)
self.f_infinity_fit = popt
def fit_boltzmann(self):
max_f0 = float(max(self.f_zeros))
min_f0 = 0.1 # float(min(self.f_zeros))
mean_int = float(np.mean(self.stimulus_value))
total_increase = max_f0 - min_f0
total_change_int = max(self.stimulus_value) - min(self.stimulus_value)
start_k = float((total_increase / total_change_int * 4) / max_f0)
popt, pcov = curve_fit(fu.full_boltzmann, self.stimulus_value, self.f_zeros,
p0=(max_f0, min_f0, start_k, mean_int),
maxfev=10000, bounds=([0, 0, -np.inf, -np.inf], [5000, 1, np.inf, np.inf]))
self.boltzmann_fit_vars = popt
def __calculate_f_baseline__(self, time, frequency, buffer=0.025):
stim_start = self.cell_data.get_stimulus_start() - time[0]
sampling_interval = self.cell_data.get_sampling_interval()
if stim_start < 0.1:
warn("FICurve:__calculate_f_baseline__(): Quite short delay at the start.")
start_idx = 0
end_idx = int((stim_start-buffer)/sampling_interval)
f_baseline = np.mean(frequency[start_idx:end_idx])
return f_baseline
def __calculate_f_zero__(self, time, frequency, peak_buffer_percent=0.05, buffer=0.025):
stimulus_start = self.cell_data.get_stimulus_start() - time[0] # time start is generally != 0 and != delay
sampling_interval = self.cell_data.get_sampling_interval()
freq_before = frequency[0:int((stimulus_start - buffer) / sampling_interval)]
min_before = min(freq_before)
max_before = max(freq_before)
mean_before = np.mean(freq_before)
# time where the f-zero is searched in
start_idx = int((stimulus_start-0.1*buffer) / sampling_interval)
end_idx = int((stimulus_start + buffer) / sampling_interval)
min_during_start_of_stim = min(frequency[start_idx:end_idx])
max_during_start_of_stim = max(frequency[start_idx:end_idx])
if abs(mean_before-min_during_start_of_stim) > abs(max_during_start_of_stim-mean_before):
f_zero = min_during_start_of_stim
else:
f_zero = max_during_start_of_stim
peak_buffer = (max_before - min_before) * peak_buffer_percent
if min_before - peak_buffer <= f_zero <= max_before + peak_buffer:
end_idx = start_idx + int((end_idx-start_idx)/2)
f_zero = np.mean(frequency[start_idx:end_idx])
return f_zero
# start_idx = int(stimulus_start / sampling_interval)
# end_idx = int((stimulus_start + buffer*2) / sampling_interval)
#
# freq_before = frequency[start_idx-(int(length_of_mean/sampling_interval)):start_idx]
# fb_mean = np.mean(freq_before)
# fb_std = np.std(freq_before)
#
# peak_frequency = fb_mean
# count = 0
# for i in range(start_idx + 1, end_idx):
# if fb_mean-3*fb_std <= frequency[i] <= fb_mean+3*fb_std:
# continue
#
# if abs(frequency[i] - fb_mean) > abs(peak_frequency - fb_mean):
# peak_frequency = frequency[i]
# count += 1
# return peak_frequency
def __calculate_f_infinity__(self, time, frequency, length=0.1, buffer=0.025):
stimulus_end_time = self.cell_data.get_stimulus_start() + self.cell_data.get_stimulus_duration() - time[0]
start_idx = int((stimulus_end_time - length - buffer) / self.cell_data.get_sampling_interval())
end_idx = int((stimulus_end_time - buffer) / self.cell_data.get_sampling_interval())
# TODO add way to plot detected f_zero, f_inf, f_base. With detection of remaining slope?
# x = np.arange(start_idx, end_idx, 1) # time[start_idx:end_idx]
# slope, intercept, r_value, p_value, std_err = linregress(x, frequency[start_idx:end_idx])
# if p_value < 0.0001:
# plt.title("significant slope: {:.2f}, p: {:.5f}, r: {:.5f}".format(slope, p_value, r_value))
# plt.plot(x, [i*slope + intercept for i in x], color="black")
#
#
# plt.plot((start_idx, end_idx), (np.mean(frequency[start_idx:end_idx]), np.mean(frequency[start_idx:end_idx])), label="f_inf")
# plt.legend()
# plt.show()
# plt.close()
return np.mean(frequency[start_idx:end_idx])
def get_f_zero_inverse_at_frequency(self, frequency):
b_vars = self.boltzmann_fit_vars
return fu.inverse_full_boltzmann(frequency, b_vars[0], b_vars[1], b_vars[2], b_vars[3])
def get_f_infinity_frequency_at_stimulus_value(self, stimulus_value):
infty_vars = self.f_infinity_fit
return fu.clipped_line(stimulus_value, infty_vars[0], infty_vars[1])
def get_f_infinity_slope(self):
return self.f_infinity_fit[0]
def get_fi_curve_slope_at(self, stimulus_value):
fit_vars = self.boltzmann_fit_vars
return fu.derivative_full_boltzmann(stimulus_value, fit_vars[0], fit_vars[1], fit_vars[2], fit_vars[3])
def get_fi_curve_slope_of_straight(self):
fit_vars = self.boltzmann_fit_vars
return fu.full_boltzmann_straight_slope(fit_vars[0], fit_vars[1], fit_vars[2], fit_vars[3])
def get_f_zero_and_f_inf_intersection(self):
x_values = np.arange(min(self.stimulus_value), max(self.stimulus_value), 0.0001)
fit_vars = self.boltzmann_fit_vars
f_zero = fu.full_boltzmann(x_values, fit_vars[0], fit_vars[1], fit_vars[2], fit_vars[3])
f_inf = fu.clipped_line(x_values, self.f_infinity_fit[0], self.f_infinity_fit[1])
intersection_indicies = np.argwhere(np.diff(np.sign(f_zero - f_inf))).flatten()
# print("fi-curve calc intersection:", intersection_indicies, x_values[intersection_indicies])
if len(intersection_indicies) > 1:
f_baseline = np.median(self.f_baselines)
best_dist = np.inf
best_idx = -1
for idx in intersection_indicies:
dist = abs(fu.clipped_line(x_values[idx], self.f_infinity_fit[0], self.f_infinity_fit[1]) - f_baseline)
if dist < best_dist:
best_dist = dist
best_idx = idx
return x_values[best_idx]
elif len(intersection_indicies) == 0:
raise ValueError("No intersection found!")
else:
return x_values[intersection_indicies[0]]
def get_fi_curve_slope_at_f_zero_intersection(self):
x = self.get_f_zero_and_f_inf_intersection()
fit_vars = self.boltzmann_fit_vars
return fu.derivative_full_boltzmann(x, fit_vars[0], fit_vars[1], fit_vars[2], fit_vars[3])
def plot_fi_curve(self, savepath: str = None, comp_f_baselines=None, comp_f_zeros=None, comp_f_infs=None):
min_x = min(self.stimulus_value)
max_x = max(self.stimulus_value)
step = (max_x - min_x) / 5000
x_values = np.arange(min_x, max_x, step)
plt.plot(self.stimulus_value, self.f_baselines, color='blue', label='f_base')
if comp_f_baselines is not None:
plt.plot(self.stimulus_value, comp_f_baselines, 'o', color='skyblue', label='comp_values base')
plt.plot(self.stimulus_value, self.f_infinities, 'o', color='green', label='f_inf')
plt.plot(x_values, [fu.clipped_line(x, self.f_infinity_fit[0], self.f_infinity_fit[1]) for x in x_values],
color='darkgreen', label='f_inf_fit')
if comp_f_infs is not None:
plt.plot(self.stimulus_value, comp_f_infs, 'o', color='lime', label='comp values f_inf')
plt.plot(self.stimulus_value, self.f_zeros, 'o', color='orange', label='f_zero')
popt = self.boltzmann_fit_vars
plt.plot(x_values, [fu.full_boltzmann(x, popt[0], popt[1], popt[2], popt[3]) for x in x_values],
color='red', label='f_0_fit')
if comp_f_zeros is not None:
plt.plot(self.stimulus_value, comp_f_zeros, 'o', color='wheat', label='comp_values f_zero')
plt.legend()
plt.ylabel("Frequency [Hz]")
if self.using_contrast:
plt.xlabel("Stimulus contrast")
else:
plt.xlabel("Stimulus intensity [mv]")
if savepath is None:
plt.show()
else:
plt.savefig(savepath + "fi_curve.png")
plt.close()
def plot_f_point_detections(self):
mean_frequencies = np.array(self.cell_data.get_mean_isi_frequencies())
time_axes = self.cell_data.get_time_axes_mean_frequencies()
for i in range(len(mean_frequencies)):
fig, axes = plt.subplots(1, 1, sharex="all")
axes.plot(time_axes[i], mean_frequencies[i], label="voltage")
axes.plot((time_axes[i][0],time_axes[i][-1]), (self.f_zeros[i], self.f_zeros[i]), label="f_zero")
axes.plot((time_axes[i][0],time_axes[i][-1]), (self.f_infinities[i], self.f_infinities[i]), '--', label="f_inf")
axes.plot((time_axes[i][0],time_axes[i][-1]), (self.f_baselines[i], self.f_baselines[i]), label="f_base")
axes.set_title(str(self.stimulus_value[i]))
plt.legend()
plt.show()