PSICS User Guide
PSICS computes the behavior of neuron models in which the ion channels behave deterministically or stochastically.
The simplest way to run it is from the command line from a Unix shell or a Windows MSDOS prompt. It reads an input model, runs the calculation and writes the output to one or more files in a new folder with a name based on the name of the model file. All its behavior is controlled by the input model specification which is provided as as a collection of XML files for the various components of the model (morphology, channels, overall control etc).
This guide first explains how to install and run PSICS, then provides background information on the processing steps that occur when a model is run and the numerical methods used to implement the calculations. The following section explains how to set up a model and should be read in conjunction with the formats reference which explains all the available structures and options. The remaining sections cover the use of the standalone computational core for multiprocessor systems and custers and summarise some external software applications that can be useful for setting up or analysing models. Finally, the bibliography points to papers and other resources concerned with numerical methods related to those used in PSICS.
As well as the formats reference, there are a range of example models which can be used to learn about psics by running it or as a starting point for defining new models.
- Installation and getting started - how to install and run PSICS on Linux, Windows and MacOSX
- Model processing - the processing stages a model goes through and some of the main parameters that control them
- Numerical methods - the methods used for continuous and stochastic calculations including the most useful parameters for controlling their operation
- Model specification - how to define and modify models
- Native executable - running the computational core on its own
- Utilities - external resources that may be useful in conjunction with PSICS
- Bibliography - mainly numerical methods for stochastic systems