CSPRC is a MATLAB toolbox for estimating the infinitesimal phase response curve (PRC) by using Compressive Sensing (CS) algorithms. For more information about the methodology, please check out our paper in Reference. This software is licensed under GPL 3.0 License.
Currently the main part of this toolbox relies on the implementations of CS algorithms written by other great people, which are
- l1-MAGIC (by Justin Romberg)
- L1 Homotopy (by Salman Asif)
In particulary, you need to have two MATLAB functions working, l1eq_pd and DS_homotopy_function.
- Download l1-MAGIC and L1 Homotopy, and add them in your MATLAB path.
- Add CSPRC toolbox directory in your MATLAB path.
- Profit!
The basic workflow is
- Create the estimation data from the single cell recording data (
make_PRC_data). - Create an estimator object (
csprc). - Run a cross-validation test to find the best estimation parameter (
xvalidate). - Estimatie the PRC with the found parameter (
csprc.evaluate).
Take a look at the example, demo_estimation_HH.m, which shows the workflow in more detail.