Genome-scale Integrative Modelling of Gene Expression and Metabolic Networks

UoM administered thesis: Phd

  • Authors:
  • Delali Adiamah


The elucidation of molecular function of proteins encoded by genes is a major challenge in biology today. Genes regulate the amount of proteins (enzymes) needed to catalyse a metabolic reaction. There are several works on either the modelling of gene expression or metabolic network. However, an integrative model of both is not well understood and researched. The integration of both gene expression and metabolic network could increase our understanding of cellular functions and aid in analysing the effects of genes on metabolism.It is now possible to build genome-scale models of cellular processes due to the availability of high-throughput genomic, metabolic and fluxomic data along with thermodynamic information. Integrating biological information at various layers into metabolic models could also improve the robustness of models for in silico analysis.In this study, we provide a software tool for the in silico reconstruction of genome-scale integrative models of gene expression and metabolic network from relevant database(s) and previously existing stoichiometric models with automatic generation of kinetic equations of all reactions involved. To reduce computational complexity, compartmentalisation of the cell as well as enzyme inhibition is assumed to play a negligible role in metabolic function. Obtaining kinetic parameters needed to fully define and characterise kinetic models still remains a challenge in systems biology. Parameters are either not available in literature or unobtainable in the lab. Consequently, there have been numerous methods developed to predict biological behaviour that do not require the use of detailed kinetic parameters as well as techniques for estimation of parameter values based on experimental data. We present an algorithm for estimating kinetic parameters which uses fluxes and metabolites to constrain values. Our results show that our genetic algorithm is able to find parameters that fit a given data set and predict new biological states without having to re-estimate kinetic parameters.


Original languageEnglish
Awarding Institution
Award date1 Aug 2012