Ions & Channels

Ions & Channels#

The electrical behavior of a neuron comes from ion channels — pores that let specific ion species flow across the membrane. braincell models these as two cooperating layers: ion species (braincell.ion) own a reversal potential and (optionally) concentration dynamics, and channels (braincell.channel) carry current driven by that reversal potential.

Ion species#

An ion species tracks the reversal potential E and, in detailed models, the intra/extracellular concentration. The families in braincell.ion:

Class family

Behavior

*Fixed (e.g. SodiumFixed, PotassiumFixed, CalciumFixed)

constant reversal potential E — the simplest, fastest choice.

CalciumDetailed, CalciumFirstOrder

calcium with concentration dynamics (buffering, pumps, diffusion), so E follows the Nernst equation as concentration changes.

*InitNernst

reversal potential initialized from the Nernst equation.

 
import brainunit as u
import braincell

# constant sodium reversal at +50 mV
na = braincell.ion.SodiumFixed(size, E=50. * u.mV)

# calcium with first-order concentration decay
ca = braincell.ion.CalciumDetailed(size, C_rest=5e-5 * u.mM,
                                   tau=10. * u.ms, d=0.5 * u.um)

Channels#

A channel produces a current as a function of voltage (and possibly an ion concentration). braincell.channel ships a large, literature-derived library, named by <current>_<source><year> convention:

Group

Examples

Sodium

Na_HH1952, Na_Ba2002

Potassium (delayed rectifier, A-type, M-type, Kv, Kir, …)

K_HH1952, KDR_Ba2002

Calcium (L/N/T/P-type, Cav families)

CaL_IS2008, CaT_HM1992

Potassium–calcium (Ca-activated K)

AHP_De1994, Kca1p1_*, Kca2p2_*

Hyperpolarization-activated (H-current)

HCN_HM1992

Leak

IL, LeakageChannel

See the braincell.channel module reference for the complete list.

How ions and channels connect#

Most channels carry a specific ion’s current, so they are added to that ion. The ion provides the reversal potential the channel’s current depends on:

na = braincell.ion.SodiumFixed(size, E=50. * u.mV)
na.add(INa=braincell.channel.Na_HH1952(size))   # INa uses na.E

For a multi-compartment Cell you express the same relationship declaratively, naming the channel as a string in a Channel (see Mechanisms).

Channels that depend on two ions: MixIons#

Some channels (calcium-activated potassium currents, for example) depend on two ion species at once — a potassium reversal potential and an intracellular calcium concentration. braincell.MixIons bundles the relevant ions so such a channel can read both:

k  = braincell.ion.PotassiumFixed(size, E=-90. * u.mV)
ca = braincell.ion.CalciumDetailed(size, ...)

kca = braincell.MixIons(k, ca)
kca.add(IAHP=braincell.channel.AHP_De1994(size, g_max=0.3 * (u.mS / u.cm**2)))

The convenience function braincell.mix_ions() builds the same combination.

The registry#

Channels and ions self-register at import time via the @register_channel / @register_ion / @register_synapse decorators in braincell.mech. That registry is what lets you refer to a channel by string name (mech.Channel("Na_Ba2002", ...)). To add your own, see Extending braincell.

See also#