NEST implementation of the astrocyte_lr_1994 model¶
The purpose of this notebook is to provide a reference solution for the the astrocyte_lr_1994 model in NEST. The model is based on Li, Y. X., & Rinzel, J. (1994), De Young, G. W., & Keizer, J. (1992), and Nadkarni, S., & Jung, P. (2003).
This notebook demonstrates how the dynamics of astrocyte_lr_1994 is implemented, and generates a recording of the dynamics (test_astrocyte.dat). This recording serves as a reference for the verification of astrocyte_lr_1994 using test_astrocyte.py in the PyNEST tests.
The problem¶
The equations governing the evolution of the astrocyte model are
\[\begin{split}\left\lbrace\begin{array}{rcl}
\frac{d[\mathrm{Ca^{2+}}](t)}{dt} &=& J_{\mathrm{channel}}(t) - J_{\mathrm{pump}}(t) + J_{\mathrm{leak}}(t) \\
J_{\mathrm{channel}}(t) &=& \mathrm{r_{ER,cyt}} \cdot \mathrm{v_{IP_3R}} \cdot m_{\infty}(t)^3 \cdot n_{\infty}(t)^3\cdot h_{\mathrm{\mathrm{IP_3R}}}(t)^3 \cdot \left([\mathrm{Ca^{2+}}]_{\mathrm{ER}} - [\mathrm{Ca^{2+}}](t)\right) \\
J_{\mathrm{pump},k}(t) &=& \frac{\mathrm{v_{SERCA}} \cdot [\mathrm{Ca^{2+}}](t)^2}{K_{\mathrm{m,SERCA}}^2+[\mathrm{Ca^{2+}}](t)^2} \\
J_{\mathrm{leak}}(t) &=& \mathrm{r_{ER,cyt}} \cdot \mathrm{v_L} \cdot \left( [\mathrm{Ca^{2+}}]_{\mathrm{ER}}(t) - [\mathrm{Ca^{2+}}](t) \right) \\
m_{\infty}(t) &=& \frac{[\mathrm{IP_3}](t)}{[\mathrm{IP_3}](t) + \mathrm{K_{d,IP_3,1}}} \\
n_{\infty}(t) &=& \frac{[\mathrm{Ca^{2+}}](t)}{[\mathrm{Ca^{2+}}](t) + \mathrm{K_{d,act}}} \\
[\mathrm{Ca^{2+}}]_{\mathrm{ER}}(t) &=& \frac{[\mathrm{Ca^{2+}}]_{\mathrm{tot}}-[\mathrm{Ca^{2+}}](t)}{\mathrm{r_{ER,cyt}}} \\
\frac{dh_{\mathrm{IP_3}}(t)}{dt} &=& \alpha_{h_{\mathrm{IP_3}}}(t) \cdot (1-h_{\mathrm{IP_3}}(t)) - \beta_{h_{\mathrm{IP_3}}}(t) \cdot h_{\mathrm{IP_3}} (t) \\
\alpha_{h_{\mathrm{IP_3}}}(t) &=& \mathrm{k_{IP_3R}} \cdot \mathrm{K_{d,inh}} \cdot \frac{[\mathrm{IP_3}](t) + \mathrm{K_{d,IP_3,1}}}{[\mathrm{IP_3}](t)+\mathrm{K_{d,IP_3,2}}} \\
\beta_{h_{\mathrm{IP_3}}}(t) &=& \mathrm{k_{IP_3R}} \cdot [\mathrm{Ca^{2+}}](t) \\
\frac{d[\mathrm{IP_3}](t)}{dt} &=& \frac{[\mathrm{IP_3}]^{*} - [\mathrm{IP_3}](t)}{\tau_\mathrm{IP_3}} + \Delta_\mathrm{IP3} \cdot J_\mathrm{syn}(t)
\end{array}\right.\end{split}\]
where \([\mathrm{IP_3}]\), \([\mathrm{Ca^{2+}}]\), and \(h_{\mathrm{IP_3}}\) are the state variables of interest.
Technical details and requirements¶
Implementation of the functions¶
The reference solution is implemented through scipy.
Requirements¶
To run this notebook, you need:
[1]:
import numpy as np
from scipy.integrate import odeint
Reference solution¶
Right hand side function¶
[2]:
def rhs(y, _, p):
"""
Implementation of astrocyte dynamics.
Parameters
----------
y : list
Vector containing the state variables [IP3, Ca_astro, h_IP3R]
_ : unused var
p : Params instance
Object containing the astrocyte parameters.
Returns
-------
dCa_astro : double
Derivative of Ca_astro
dIP3 : double
Derivative of IP3
dh_IP3R : double
Derivative of h_IP3R
"""
IP3 = y[0]
Ca_astro = y[1]
h_IP3R = y[2]
Ca_astro = max(0, min(Ca_astro, p.Ca_tot * (1 + p.ratio_ER_cyt)))
alpha = p.k_IP3R * p.Kd_inh * (IP3 + p.Kd_IP3_1) / (IP3 + p.Kd_IP3_2)
beta = p.k_IP3R * Ca_astro
Ca_ER = (p.Ca_tot - Ca_astro) / p.ratio_ER_cyt
m_inf = IP3 / (IP3 + p.Kd_IP3_1)
n_inf = Ca_astro / (Ca_astro + p.Kd_act)
J_ch = p.ratio_ER_cyt * p.rate_IP3R * ((m_inf * n_inf * h_IP3R) ** 3) * (Ca_ER - Ca_astro)
J_pump = p.rate_SERCA * (Ca_astro**2) / (p.Km_SERCA**2 + Ca_astro**2)
J_leak = p.ratio_ER_cyt * p.rate_L * (Ca_ER - Ca_astro)
dCa_astro = J_ch - J_pump + J_leak
dIP3 = (p.IP3_0 - IP3) / p.tau_IP3
dh_IP3R = alpha * (1 - h_IP3R) - beta * h_IP3R
return dIP3, dCa_astro, dh_IP3R
Complete model¶
[3]:
def scipy_astrocyte(p, f, simtime, dt, spk_ts, spk_ws):
"""
Complete astrocyte model using scipy `odeint` solver.
Parameters
----------
p : Params instance
Object containing the astrocyte parameters.
f : function
Right-hand side function
simtime : double
Duration of the simulation (will run between
0 and tmax)
dt : double
Time increment.
spk_ts, spk_ws : list
Times and weights of spike input to the astrocyte
Returns
-------
t : list
Times at which the astrocyte state was evaluated.
y : list
State values associated to the times in `t`
fos : list
List of dictionaries containing additional output
information from `odeint`
"""
t = np.arange(0, simtime, dt) # time axis
n = len(t)
y = np.zeros((n, 3)) # state variables: IP3, Ca_astro, h_IP3R
# for the state variables, assign the same initial values as in test_astrocyte.py
y[0, 0] = 1.0
y[0, 1] = 1.0
y[0, 2] = 1.0
fos = [] # full output dict from odeint()
delta_ip3 = 5.0 # parameter determining the increase in IP3 induced by synaptic input
# update time-step by time-step
for k in range(1, n):
# solve ODE from t_k-1 to t_k
d, fo = odeint(f, y[k - 1, :], t[k - 1 : k + 1], (p,), full_output=True)
y[k, :] = d[1, :]
# apply synaptic inputs (spikes)
if t[k] in spk_ts:
y[k, 0] += delta_ip3 * spk_ws[spk_ts.index(t[k])]
fos.append(fo)
return t, y, fos
Parameters¶
[4]:
astro_param = {
"Ca_tot": 2.0,
"IP3_0": 0.16,
"Kd_act": 0.08234,
"Kd_inh": 1.049,
"Kd_IP3_1": 0.13,
"Kd_IP3_2": 0.9434,
"Km_SERCA": 0.1,
"ratio_ER_cyt": 0.185,
"delta_IP3": 5.0,
"k_IP3R": 0.0002,
"rate_L": 0.00011,
"tau_IP3": 7142.0,
"rate_IP3R": 0.006,
"rate_SERCA": 0.0009,
}
class Params:
"""
Class giving access to the astrocyte parameters.
"""
def __init__(self):
self.params = astro_param
self.Ca_tot = astro_param["Ca_tot"]
self.IP3_0 = astro_param["IP3_0"]
self.Kd_act = astro_param["Kd_act"]
self.Kd_inh = astro_param["Kd_inh"]
self.Kd_IP3_1 = astro_param["Kd_IP3_1"]
self.Kd_IP3_2 = astro_param["Kd_IP3_2"]
self.Km_SERCA = astro_param["Km_SERCA"]
self.ratio_ER_cyt = astro_param["ratio_ER_cyt"]
self.delta_IP3 = astro_param["delta_IP3"]
self.k_IP3R = astro_param["k_IP3R"]
self.rate_L = astro_param["rate_L"]
self.tau_IP3 = astro_param["tau_IP3"]
self.rate_IP3R = astro_param["rate_IP3R"]
self.rate_SERCA = astro_param["rate_SERCA"]
p = Params()
Simulate the implementation¶
[5]:
# Parameters for the simulation
simtime = 100.0
resolution = 0.1
spike_times = [10.0]
spike_weights = [1.0]
# Simulate and get the recording
t, y, fos = scipy_astrocyte(p, rhs, simtime, resolution, spike_times, spike_weights)
data = np.concatenate((np.array([t]).T, y), axis=1)
# Save the recording (excluding the initial values)
np.savetxt(
"test_astrocyte.dat",
data[1:, :],
header="""\ntest_astrocyte.dat\n
This file is part of NEST.\n
This .dat file contains the recordings of the state variables of an
astrocyte, simulated using the ODEINT solver of SciPy, with the
implementation detailed in
``doc/htmldoc/model_details/astrocyte_model_implementation.ipynb``.
This data is used as reference for the tests in ``test_astrocyte.py``.\n
Times IP3 Ca_astro h_IP3R""",
)
License¶
This file is part of NEST. Copyright (C) 2004 The NEST Initiative
NEST is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 2 of the License, or (at your option) any later version.
NEST is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.