Skip to content

An application for fitting BioNetGen models using a genetic algorithm

License

Notifications You must be signed in to change notification settings

RuleWorld/BioNetFit

Repository files navigation

Notice

This Perl version of BioNetFit is deprecated, but remains available for archival purposes. The new version of BioNetFit is available at https://github.com/lanl/PyBNF.

BioNetFit

BioNetFit is a general-purpose fitting program for rule-based models written in the BioNetGen language.

Features

  • BioNetFit implements a genetic algorithm for optimization
  • BioNetFit can be used on a stand-alone computer or a Linux-based cluster
  • BioNetFit is compatible with the various simulation capabilities provided by BioNetGen (generate-first deterministic, stochastic and hybrid methods and network-free stochastic methods) can be used with BioNetFit.
  • BioNetFit can perform a global fit, meaning that it can consider multiple datasets in the course of a fitting run. Preprocessing of data (for example, normalization of the average of the measurement values in each dataset) is advised when considering multiple datasets.
  • BioNetFit can consider steady-state dose-response data and/or time-series data.
  • BioNetFit supports bootstrapping to characterize parameter confidence intervals (not fully tested, use at your own peril)
    • the parameter bootstrap_chi is the maximum objective function value (numeric) required for accepting a particular bootstrapped fitting run
    • the parameter bootstrap takes an integer defining the number of bootstrapped fitting runs to perform

Support

A user manual is present in the repository

Additional support is generally available by contacting the BioNetGen developers

About

An application for fitting BioNetGen models using a genetic algorithm

Resources

License

Stars

Watchers

Forks

Packages

No packages published