Mathematics for Computational Neuroscience

Hosted by Speaker: Alex Williams

1. Membrane Potential Introduction

This lesson introduces the membrane potential equation

What is Membrane Potential?

The membrane potential ($V_m$) is the voltage difference across a neuron’s membrane, defined as $V_m = V_i - V_e$, where $V_i$ is the intracellular potential and $V_e$ is the extracellular potential.

  • It’s measured in millivolts (mV), with the inside of the cell typically being negative relative to the outside (e.g., ~-70 mV at rest).
  • This potential arises from the asymmetric distribution of ions like Sodium (Na+), Potassium (K+), and Chloride (Cl-) across the membrane, a state maintained by active ion pumps.

Electrical Properties of the Membrane

To model the cell membrane, we can abstract its physical properties into an equivalent electrical circuit:

  1. Capacitor (insulator): The lipid bilayer separates charges, acting like a capacitor that can store charge. This is crucial for generating action potentials.
  2. Resistors (conductors): Ion channels allow ions to move across the membrane, functioning as resistors.
  3. Voltage Source (Battery): The concentration gradients of ions create an electrochemical potential, which acts as a battery.

This leads to the simplest model of a cell membrane patch as an RC circuit:

---
title: Simplest Cellular Membrane Circuit
---
graph TD
    subgraph "Inside (V_i)"
        A --- C
    end
    subgraph "Outside (V_e)"
        B --- D
    end

    subgraph "Membrane"
        C -- C_m --> D
        C -- R_m --> E
        E -- E_L --> D
    end

    A -- V_m --- B

    linkStyle 3 stroke-width:0px,fill:none;

Governing Equations

  • Nernst equation: Calculates the equilibrium potential for a single ion species.
  • Goldman equation: An extension of the Nernst equation for multiple ion species, used to calculate the resting membrane potential.

2. The Membrane Equation (Passive Neuron)

3. Separation of Variables (Solving Passive Membrane)

Solving the passive membrane equation

4. Injecting Current Into a Passive Membrane

Injecting current into a passive membrane

5. Response to a Current Step

Response to injected current

6. Numerically Solving the Membrane Equation

Explains the logic behind dealing with more complex currents by solving the membrane equation numerically.

7. Intro to Conductance-Based Models

Introducing voltage-dependent ion channels into the passive membrane

8. Hodgkin Huxley Channel Models

Introducing Hodgkin & Huxley’s voltage dependent ion channel models, with emphasis on the sodium conductance

9. Hodgkin-Huxley Squid Axon Model

Introducing the classical Hodgkin & Huxley squid axon model with sodium and potassium conductances

10. Multi-Compartment Conductance-Based Models

This lesson extends the conductance-based model equation to multiple neuronal compartments, taking more complex morphology into account.

References

INCF Basic Mathematics for Computational Neuroscience




    Enjoy Reading This Article?

    Here are some more articles you might like to read next:

  • In memorial of David C.Marr
  • Neuromatch NeuroAI course
  • How to start Computational Neuroscience
  • Harvard CS197 AI Research Experiences
  • Learning Resources