Molecular evolution & bioinformatics
Modelling the major evolutionary transitions
My research looks at an enigma in the major evolutionary transitions: the problem of why an individual would invest resources for some genetic stake in an offspring rather than invest in its own clonal offspring. The problem is tackled with models of resource allocation strategy, studying invasion of different strategies done with computer simulation models. The results are relevant to Artificial Life, Social Evolution and Adaptive Dynamics.
Modelling gene regulation networks
My research is mainly concerned with theoretical modeling and analysis of genetic regulatory networks (GNR) in organisms and especially in the bacterium S. aureus. Basically, a GNR outlines the functioning of an organism on the cellular level which involves interactions between DNA, RNA and proteins. Such interactions play a key role in the life and phenotype of organisms since they dictate which genes should be expressed and under what circumstances.
Compositional structures in non-coding DNA
How are nucleotides distributed along genomes? Is there a functional significance to the organization of bases along noncoding DNA, and do compositional structures reflect fundamental laws underlying evolutionary processes? Several decades ago, a series of seminar experiments uncovered surprisingly long sequences of DNA across a variety of species that have a remarkably homogeneous composition of bases. On the basis of these results, the ``isochore'' theory was proposed to account for the structure of the genomes of all warm-blooded vertebrates. In 2001, the sequencing of the complete human genome paved the way for a sequence-based reinspection of the isochore theory, which has been the focus of our work.
Bayesian networks for bioinformatics
A number of bioinformatics applications have been developed within the group. A number of these were developed during a recent BBSRC project on Protein Function Prediction using uncertainty, worked on by Chris Needham and Andy Bulpitt. Earlier work related to proteins can be found here.
Machine learning approach to reverse engineering gene regulatory networks from gene expression data is also underway. Information about this can be found on Dr Needham's page introducing Learning gene regulatory networks in Arabidopsis.