Computer Science and
     Software Engineering

Computer Science and Software Engineering

HONS 09/06

Stochastic modelling of horizontal gene transfer: The effect of mutator heterogeneity on sequence fixation

Leighton Turner
Department of Computer Science
University of Canterbury

Abstract

The evolutionary and environmental consequences of horizontal gene transfer (HGT) is illustrated by the global spread of antibiotic resistance to pathogenic bacteria in under 50 years. Another topical question is whether genes from transgenic organisms will spread into unintended hosts such as soil bacteria. Why was the first phenomenon not anticipated and the second issue uncertain? Current mathematical models of HGT do not have sufficient predictive power. While they conclude that genes can persist in populations for long times even if the gene confers a slightly negative selection coefficient, they also assume every individual is equally capable at recombining foreign DNA. This assumption is clearly incorrect, but does that matter? As an analogy to heterogeneous epidemiological models where certain individuals – superspreaders – are better at spreading disease than others, this project considers HGT in a heterogeneous population of mutator phenotypes where some individuals are superacceptors of foreign DNA. Using a stochastic simulation, the qualitative affect of mutator heterogeneity on sequence fixation rates was determined. As the distribution of mutator phenotypes became more heterogeneous, the time required for a population to fix a gene increased and became more variable. The theoretical validity of this model is discussed and comments made about the difficulty of confirming this model in vivo. For the immediate future, experimentation into the complex interaction between population heterogeneity and HGT seems best left too in silico studies.