Clive Bowsher

Research in Mathematical Systems Biology

My ongoing research is in the field of mathematical and theoretical systems biology. I use approaches from probability and statistics to advance understanding of the dynamics and functioning of living cells. At the level of single cells, it is now widely appreciated that stochasticity has important phenotypic effects and a probabilistic approach is therefore often needed.

Recent work (Nucleic Acids Research, with Margaritis Voliotis, pdf) examines the control of fluctuations in gene expression by negative feedback, comparing transcriptional and small RNA-mediated autoregulatory mechanisms. The results also highlight the importance of separately considering effects on both the mean and the variance when comparing two systems, rather than relying on a single summary measure such as the coefficient of variation.

Cells use stochastic biochemistry to sense both intracellular and environmental fluctuations, and to convey essential information generated by these changes to downstream effectors. A current focus is on understanding how such information is encoded and propagated by biomolecular networks.

In "Identifying Sources of Variation and the Flow of Information in Biochemical Networks" (PNAS, with Peter Swain) we provide a new framework for understanding sources of variation between cells and for quantifying the effects of information flow on cellular heterogeneity. To understand how cells control and exploit biochemical fluctuations, we must identify the sources of stochasticity, quantify their effects, and distinguish informative variation from confounding `noise'. We present an analysis that allows fluctuations of biochemical networks to be decomposed into multiple components, gives conditions for the design of experimental reporters to measure all components, and provides a technique to evaluate the magnitude of these components for a given model. Further, we identify a particular component of variation that can be used to quantify the efficacy of information flow through a biochemical network.

In "Information Processing by Biochemical Networks: A Dynamic Approach" (SI) (Royal Society Interface) I introduced an approach grounded in biochemical, mass action kinetics and based on dynamic conditional independences between species trajectories. (A 'species trajectory' is the time course of the number of molecules of a particular type of biomolecule). By deriving dynamic conditional independences, we can identify the species trajectories that fully encode the relevant information and trace the sequential process of information transfer through the biochemical network. The approach is applicable to a wide class of stochastic dynamics. (ICSB presentation).

The underlying mathematics needed for the analysis of conditional independences and a graphical treatment of biochemical kinetics is provided in "Stochastic Kinetic Models: Dynamic Independence, Modularity and Graphs" (Annals of Statistics). This was the first paper to view the modular architecture of biomolecular networks in terms of the independence properties of the dynamics.

MIDIA is a powerful algorithm that allows for automatation of the approaches described above. In "Automated analysis of information processing, kinetic independence and modular architecture in biochemical networks using MIDIA" (SI) (Bioinformatics) the algorithm is explained and is made available to the biological community as an extensible software package in the R language (download MIDIA software). MIDIA computes exact network decompositions based on dynamic independence properties of the modules and identifies important biochemical intermediaries that result in the overlap of modules.

The book chapter "Dynamic Molecular Networks and Mechanisms in the Biosciences: A Statistical Framework" provides a review of the area from the perspective of statistics and causal modelling.

Peter Swain (lab pages)
Margaritis Voliotis

Related conference presentations

Physics Meets Biology, Oxford, 2012
Applied Probability in Modern Biology, Royal Statistical Society Meeting, 2012 (invited speaker)
Time for Causality Conference, University of Bristol, 2012 (invited speaker)
Conference on Stochastic Systems Biology, Monte Verita Switzerland, 2011 (invited speaker)
11th International Conference on Systems Biology, ICSB 2010, Edinburgh
Wellcome Trust-Cold Spring Harbor Laboratory Meeting on Networks:Systems Biology, Hinxton, 2010
Statistics of Networks Conference, University of Bristol, 2010 (invited speaker)

Recent reviews undertaken for PLoS Computational Biology, Bioinformatics, BMC Systems Biology, Cell Communication & Signaling, Journal of the American Statistical Association, and Journal of Causal Inference

Back to Clive Bowsher's Homepage