Stochastic dynamics of genetic networks
Bioinformatics
Hybrid numerical solution of the chemical master equation
Proceedings of the 8th International Conference on Computational Methods in Systems Biology
SHAVE: stochastic hybrid analysis of markov population models
Proceedings of the 14th international conference on Hybrid systems: computation and control
Parameter identification for Markov models of biochemical reactions
CAV'11 Proceedings of the 23rd international conference on Computer aided verification
The Monte Carlo EM method for the parameter estimation of biological models
Electronic Notes in Theoretical Computer Science (ENTCS)
Hi-index | 0.00 |
The dynamics of biochemical reaction networks can be accurately described by stochastic hybrid models, where we assume that large chemical populations evolve deterministically and continuously over time while small populations change through random discrete reactions. We propose an algorithm for estimating the parameters of a given biochemical reaction network based on a stochastic hybrid model. We assume that noisy time series measurements of the chemical populations are available and follow a maximum likelihood approach to calibrate the parameters. We numerically approximate the likelihood and its derivatives for concrete values of the parameters and show that, based on this approximation, the maximization of the likelihood can be done efficiently. We substantiate the usefulness of our approach by applying it to several case studies from systems biology.