Package was developed by ciaran welsh (email@example.com) from Newcastle University, UK.
pip install PyCoTools
PyCoTools is a Python package for interfacing with COAPSI and performing modelling tasks that use COPASI as a simulation engine.
PyCoTools has three modules at the moment:
Firstly, copasiAPI is a set of python classes for interfacing with copasi via python. This is intended to
eventually provide the same modelling flexibility as copasi but currently only supports performing parameter
estimation, running deterministic time courses and running parameter scans.
Secondly the pydentify2 module uses the copasiAPI module to set up and run the profile likelihood method of
identifiability analysis using copasi (Schaber2012). Additional support is provided to automatically read
parameter estimation output files (or folders of files) from copasi and calculate identifiability around
these points in parameter space.
The PEAnalysis provides a set of classes for analysing PE data and relies heavily on pandas, numpy, scipy and matplotlib.
The code works but is still unfinished. In particular the code style is inconsistant with the style of the other classes.
This is because the other classes are new and the PEAnalysis class will be adopting the new style asap.
Other modules will be provided in the future as well as additional classes and features of the copasiAPI such as scan cisualization or steady state task.
Examples are provided for using various aspects of PyCoTools. In particular, all curated models were downloaded
using the bioservices.biomodels package and converted into copasi files. Then time course data was simulated
for any model with less than 12 species, noise was added and a parameter estimation was performed for all model
variables with very loose optimization parameters (afterall we're just demonstrating use of the parameter estimation
rather than actually estimating parameters). Finally the parameter estimations are used in a profile likelihood using pydentify2,
which is revised from the first version to rely on the custom python-api, such that the process is fully autonomous.
Any questions feel free to email me and I'll try and respond.
TODO: Brief introduction on what you do with files - including link to relevant help section.