A carregar...
Publicação
Identification of neutral biochemical network models from time series data
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1019 KB | Adobe PDF |
Orientador(es)
Resumo(s)
Background: The major difficulty in modeling biological systems from multivariate time series is the identification of parameter sets that endow a model with dynamical behaviors sufficiently similar to the experimental data. Directly related to this parameter estimation issue is the task of identifying the structure and regulation of ill-characterized systems. Both tasks are simplified if the mathematical model is canonical, i.e., if it is constructed according to strict guidelines. Results: In this report, we propose a method for the identification of admissible parameter sets of canonical S-systems from biological time series. The method is based on a Monte Carlo process that is combined with an improved version of our previous parameter optimization algorithm. The method maps the parameter space into the network space, which characterizes the connectivity among components, by creating an ensemble of decoupled S-system models that imitate the dynamical behavior of the time series with sufficient accuracy. The concept of sloppiness is revisited in the context of these S-system models with an exploration not only of different parameter sets that produce similar dynamical behaviors but also different network topologies that yield dynamical similarity. Conclusion: The proposed parameter estimation methodology was applied to actual time series data from the glycolytic pathway of the bacterium Lactococcus lactis and led to ensembles of models with different network topologies. In parallel, the parameter optimization algorithm was applied to the same dynamical data upon imposing a pre-specified network topology derived from prior biological knowledge, and the results from both strategies were compared. The results suggest that the proposed method may serve as a powerful exploration tool for testing hypotheses and the design of new experiments.
Descrição
info:eu-repo/grantAgreement/FCT/3599-PPCDT/69530/PT The authors acknowledge partial support by project DynaMo (PTDC/EEA-ACR/69530/2006; S. Vinga, PI) from the Portuguese Science Foundation (FCT) and the INESC-ID. We also would like to thank the anonymous reviewers for the constructive suggestions. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the sponsoring institutions.
Palavras-chave
S-SYSTEM MODELS POWER-LAW APPROXIMATION PARAMETER-ESTIMATION BIOLOGICAL NETWORKS STATISTICAL-METHODS IDENTIFIABILITY OPTIMIZATION DYNAMICS ROBUSTNESS ALGORITHM Structural Biology Modelling and Simulation Molecular Biology Computer Science Applications Applied Mathematics