Constructing and simulating compartmental models with active and passive dendrites

This example demonstrates how to initialize compartmental models. It creates two models with two compartments, once with active channels in the dendritic compartment and once without. It then adds and excitatory receptors with AMPA and NMDA components to both the somatic and dendritic compartment.

The output shows the evolution of the state variables present in the model.

Authors:

WAM Wybo

import nest
import matplotlib.pyplot as plt

nest.ResetKernel()

soma_params = {
    # passive parameters
    'C_m': 89.245535,           # [pF] Capacitance
    'g_C': 0.0,                 # soma has no parent
    'g_L': 8.924572508,         # [nS] Leak conductance
    'e_L': -75.0,               # [mV] leak reversal
    # ion channel params
    'gbar_Na': 4608.698576715,  # [nS] Na maximal conductance
    'e_Na': 60.,                # [mV] Na reversal
    'gbar_K': 956.112772900,    # [nS] K maximal conductance
    'e_K': -90.                 # [mV] K reversal
}

by default, active conductances are set to zero, so we don’t need to specify them explicitly

dend_params_passive = {
    # passive parameters
    'C_m': 1.929929,
    'g_C': 1.255439494,
    'g_L': 0.192992878,
    'e_L': -75.0,
}
dend_params_active = {
    # passive parameters
    'C_m': 1.929929,            # [pF] Capacitance
    'g_C': 1.255439494,         # [nS] Coupling conductance to parent (soma here)
    'g_L': 0.192992878,         # [nS] Leak conductance
    'e_L': -70.0,               # [mV] leak reversal
    # ion channel params
    'gbar_Na': 17.203212493,    # [nS] Na maximal conductance
    'e_Na': 60.,                # [mV] Na reversal
    'gbar_K': 11.887347450,     # [nS] K maximal conductance
    'e_K': -90.                 # [mV] K reversal
}

create a neuron model with a passive dendritic compartment

cm_pas = nest.Create('cm_default')
cm_pas.compartments = [
    {"parent_idx": -1, "params": soma_params},
    {"parent_idx":  0, "params": dend_params_passive}
]

create a neuron model with an active dendritic compartment

cm_act = nest.Create('cm_default')
cm_act.compartments = [
    {"parent_idx": -1, "params": soma_params},
    {"parent_idx":  0, "params": dend_params_active}
]

set spike thresholds

cm_pas.V_th = -50.
cm_act.V_th = -50.

add somatic and dendritic receptor to passive dendrite model

cm_pas.receptors = [
    {"comp_idx": 0, "receptor_type": "AMPA_NMDA"},
    {"comp_idx": 1, "receptor_type": "AMPA_NMDA"}
]
syn_idx_soma_pas = 0
syn_idx_dend_pas = 1

add somatic and dendritic receptor to active dendrite model

cm_act.receptors = [
    {"comp_idx": 0, "receptor_type": "AMPA_NMDA"},
    {"comp_idx": 1, "receptor_type": "AMPA_NMDA"}
]
syn_idx_soma_act = 0
syn_idx_dend_act = 1

create a two spike generators

sg_soma = nest.Create('spike_generator', 1, {'spike_times': [10., 13., 16.]})
sg_dend = nest.Create('spike_generator', 1, {'spike_times': [70., 73., 76.]})

connect spike generators to passive dendrite model (weight in nS)

nest.Connect(sg_soma, cm_pas, syn_spec={
    'synapse_model': 'static_synapse', 'weight': 5., 'delay': 0.5, 'receptor_type': syn_idx_soma_pas})
nest.Connect(sg_dend, cm_pas, syn_spec={
    'synapse_model': 'static_synapse', 'weight': 2., 'delay': 0.5, 'receptor_type': syn_idx_dend_pas})

connect spike generators to active dendrite model (weight in nS)

nest.Connect(sg_soma, cm_act, syn_spec={
    'synapse_model': 'static_synapse', 'weight': 5., 'delay': 0.5, 'receptor_type': syn_idx_soma_act})
nest.Connect(sg_dend, cm_act, syn_spec={
    'synapse_model': 'static_synapse', 'weight': 2., 'delay': 0.5, 'receptor_type': syn_idx_dend_act})

create multimeters to record compartment voltages and various state variables

rec_list = ['v_comp0', 'v_comp1',
            'm_Na_0', 'h_Na_0', 'n_K_0', 'm_Na_1', 'h_Na_1', 'n_K_1',
            'g_r_AN_AMPA_1', 'g_d_AN_AMPA_1', 'g_r_AN_NMDA_1', 'g_d_AN_NMDA_1']
mm_pas = nest.Create('multimeter', 1, {'record_from': rec_list, 'interval': .1})
mm_act = nest.Create('multimeter', 1, {'record_from': rec_list, 'interval': .1})

connect the multimeters to the respective neurons

nest.Connect(mm_pas, cm_pas)
nest.Connect(mm_act, cm_act)

simulate the models

nest.Simulate(160.)
res_pas = nest.GetStatus(mm_pas, 'events')[0]
res_act = nest.GetStatus(mm_act, 'events')[0]

plt.figure('voltage')

plot voltage for somatic compartment

ax_soma = plt.subplot(121)
ax_soma.plot(res_pas['times'], res_pas['v_comp0'], c='b', label='passive dend')
ax_soma.plot(res_act['times'], res_act['v_comp0'], c='r', label='active dend')
ax_soma.set_xlabel(r'$t$ (ms)')
ax_soma.set_ylabel(r'$v_{soma}$ (mV)')
ax_soma.set_ylim((-90., 40.))
ax_soma.legend(loc=0)

plot voltage for dendritic compartment

ax_dend = plt.subplot(122)
ax_dend.plot(res_pas['times'], res_pas['v_comp1'], c='b', label='passive dend')
ax_dend.plot(res_act['times'], res_act['v_comp1'], c='r', label='active dend')
ax_dend.set_xlabel(r'$t$ (ms)')
ax_dend.set_ylabel(r'$v_{dend}$ (mV)')
ax_dend.set_ylim((-90., 40.))
ax_dend.legend(loc=0)

plt.figure('channel state variables')

plot ion channel state variables for somatic compartment

ax_soma = plt.subplot(121)
ax_soma.plot(res_pas['times'], res_pas['m_Na_0'], c='b', label='m_Na passive dend')
ax_soma.plot(res_pas['times'], res_pas['h_Na_0'], c='r', label='h_Na passive dend')
ax_soma.plot(res_pas['times'], res_pas['n_K_0'], c='g', label='n_K passive dend')
ax_soma.plot(res_act['times'], res_act['m_Na_0'], c='b', ls='--', lw=2., label='m_Na active dend')
ax_soma.plot(res_act['times'], res_act['h_Na_0'], c='r', ls='--', lw=2., label='h_Na active dend')
ax_soma.plot(res_act['times'], res_act['n_K_0'], c='g', ls='--', lw=2., label='n_K active dend')
ax_soma.set_xlabel(r'$t$ (ms)')
ax_soma.set_ylabel(r'svar')
ax_soma.set_ylim((0., 1.))
ax_soma.legend(loc=0)

plot ion channel state variables for dendritic compartment

ax_dend = plt.subplot(122)
ax_dend.plot(res_pas['times'], res_pas['m_Na_1'], c='b', label='m_Na passive dend')
ax_dend.plot(res_pas['times'], res_pas['h_Na_1'], c='r', label='h_Na passive dend')
ax_dend.plot(res_pas['times'], res_pas['n_K_1'], c='g', label='n_K passive dend')
ax_dend.plot(res_act['times'], res_act['m_Na_1'], c='b', ls='--', lw=2., label='m_Na active dend')
ax_dend.plot(res_act['times'], res_act['h_Na_1'], c='r', ls='--', lw=2., label='h_Na active dend')
ax_dend.plot(res_act['times'], res_act['n_K_1'], c='g', ls='--', lw=2., label='n_K active dend')
ax_dend.set_xlabel(r'$t$ (ms)')
ax_dend.set_ylabel(r'svar')
ax_dend.set_ylim((0., 1.))
ax_dend.legend(loc=0)

plt.figure('dendritic synapse conductances')

plot synapse state variables for dendritic compartment

ax_dend = plt.gca()
ax_dend.plot(res_pas['times'], res_pas['g_r_AN_AMPA_1'] + res_pas['g_d_AN_AMPA_1'], c='b', label='AMPA passive dend')
ax_dend.plot(res_pas['times'], res_pas['g_r_AN_NMDA_1'] + res_pas['g_d_AN_NMDA_1'], c='r', label='NMDA passive dend')
ax_dend.plot(res_act['times'], res_act['g_r_AN_AMPA_1'] + res_act['g_d_AN_AMPA_1'],
             c='b', ls='--', lw=2., label='AMPA active dend')
ax_dend.plot(res_act['times'], res_act['g_r_AN_NMDA_1'] + res_act['g_d_AN_NMDA_1'],
             c='r', ls='--', lw=2., label='NMDA active dend')
ax_dend.legend(loc=0)

plt.tight_layout()
plt.show()