File Exchange

image thumbnail


version (882 KB) by Jason Sherfey
DynaSim simplifies the process of building and simulating models of dynamical systems.


Updated 10 Jul 2018

GitHub view license on GitHub

DynaSim is a MATLAB toolbox for modeling and simulating dynamical systems. It includes a library of biophysical neural models. Complete documentation is available at Please cite the DynaSim paper if you use DynaSim for your published research:

Comments and Ratings (13)

matrix tera


@Jason: I really like your idea that using existing mechanisms to build a network. And several other nice features of DynaSim, e.g. autogenerate specific solver file, varying parameter, using mex compiler to speed up simulation... Thank you for your work, it is bound to facilitate computational study of biophysical neural network!

@Quan: Thank you for your comment and observation! You are correct. There was a bug that I have fixed in the latest version on GitHub, which will be available here on the File Exchange after 12am, 02-Oct-2017.


Hi,Jason!, I am using RK4 as my solver, but when I open solve file,I notice that k1,k2,k3,k4 were updated using the same equation? I think k2 should be updated base on k1, k3 base on k2, etc...
do you think it is a mistake?


It's an INCREDIBLY powerful dynamical systems solver with an extremely easy to use interface (both via commands and GUI) that offers effortless parallelization of very computation-intensive simulations. Dynasim models are simple and intuitive to create and run, and it's trivial to enable multi-core parallelization, compiling jobs, and batch distribution. It has an extremely easy to use method built-in for multidimensional parameter sweeping, and you can submit batch jobs on a cluster (e.g. Sun Grid Engine) just by passing a flag! It is certainly the easiest to use and most powerful dynamical systems simulator I've seen for MATLAB, especially for neuroscience modeling like Hodgkin-Huxley or Integrate-and-Fire models. It probably has one of the best user interfaces of all modern computational neuroscience modeling software.

This is a really easy to use and powerful dynamical systems solver/simulator. It works especially well with models in neuroscience—it's the best Matlab toolbox for creating anything from integrate and fire models to Hodgkin-Huxley models.


reformatted description

added published article to description

converted to hyperlink

Uploaded square logo for preview

Updated image and documentation link.

Updated documentation link.

Updated the documentation link.

removed image

Added a link to Read the Docs.

MATLAB Release Compatibility
Created with R2014a
Compatible with any release
Platform Compatibility
Windows macOS Linux