AHP_De1994#
- class braincell.channel.AHP_De1994(size, n=2, g_max=Quantity(10., 'mS / cm^2'), alpha=48.0, beta=0.09, phi=1.0, name=None)#
Calcium-activated after-hyperpolarization current.
Reproduces the slow Ca2+-activated K+ (AHP) current of the thalamic reticular nucleus spindle-rhythmicity model of (Destexhe et al., 1994) [1]: a two-site calcium-binding gate with no voltage dependence, of the closed form
<closed> + n Ca_i <-> <open> (alpha, beta).\[\begin{split}\begin{aligned} I_{AHP} &= g_{\mathrm{max}} \, p^2 \, (E_K - V) \\ \frac{dp}{dt} &= \phi \left(\alpha_p \, (1 - p) - \beta_p \, p\right) \\ \alpha_p &= \alpha \, \left([\mathrm{Ca}]_i / \mathrm{mM}\right)^n \\ \beta_p &= \beta \end{aligned}\end{split}\]i.e. a Hodgkin-Huxley-style gate whose forward rate is a power law in intracellular calcium and whose backward rate is a fixed constant.
- Parameters:
size (
int|Sequence[int] |integer|Sequence[integer]) – Channel state shape.n (
Array|ndarray|bool|number|bool|int|float|complex|Quantity|Callable) – Calcium-binding exponent used inside \(\alpha_p\), default2. Independent of the gate’s conductance exponent (see Notes): changingnalters only the calcium-binding rate law, not thep ** 2factor in the current.g_max (
Array|ndarray|bool|number|bool|int|float|complex|Quantity|Callable) – Maximal conductance density, default10.0 mS/cm2.alpha (
Array|ndarray|bool|number|bool|int|float|complex|Quantity|Callable) – Forward (calcium-binding) rate coefficient, default48.0(unitsmM^-n ms^-1).beta (
Array|ndarray|bool|number|bool|int|float|complex|Quantity|Callable) – Backward (unbinding) rate, default0.09 ms^-1. See Notes: this default is three times the reference value.phi (
Array|ndarray|bool|number|bool|int|float|complex|Quantity|Callable) – Rate-scaling factor multiplying both \(\alpha_p\) and \(\beta_p\), default1.0.
Notes
No
.modfile under this repository carries aDe1994/De19filename fragment; this class has no in-repo NMODL counterpart to name here.The gate’s conductance exponent is fixed at
power=2by this class’sgatesdeclaration and is independent of thenconstructor parameter, which only exponentiates[Ca]_iinside the forward rate. Both default to2, which is why the two are easy to conflate; they do not have to agree.BrainCell’s ``beta`` default is three times the reference value. BrainCell defaults to
beta = 0.09 ms^-1, but the reference value isbeta = 0.03 ms^-1– the value used by the authors’ reference implementation, and quoted from the paper by BrainPy. (The 1994 paper itself is paywalled; only its PubMed abstract was read for this documentation pass, so the 0.03 figure could not be checked directly against the published text.) Thealpha = 48.0default is unaffected by this discrepancy: it is exactlybeta / cac ** n = 0.03 / 0.025 ** 2 = 48 mM^-2 ms^-1from the reference implementation’s ownbetaandcac, and BrainCell reproduces it exactly – evidence that the rest of the parameterisation is faithful even though thebetadefault drifted.References