By: João Carriço
At: Instituto de Investigação Interdisciplinar, Anfiteatro
The study of bacterial population genetics is of crucial importance for its understanding in the context of epidemics and outbreaks. The main genetic phenomena that drive the evolution of transmissible pathogens are mutation and recombination. Previous studies have shown that observed population genetic structure of severalÂ important human pathogens, such as Streptococcus pneumoniae and Neisseria meningitidis, can be explained using a simple evolutionary model [Fraser, PNAS 2005].
This model is based on neutral mutational drift and modulated by recombination, but incorporating the impact of epidemic transmission only for panmictic populations. Although this simple evolutionary model works well for local populations, at a ``microepidemic'' level, its predictions no longer seem to fit observed genetic relationships of large and widely distributed bacterial populations.
With the increasing volume of data obtained with sequence based typing methods, namely by Multi-Locus Sequence Typing (MLST) [maiden1998], currently the gold standard for epidemiological surveillance, a much more complex pattern emerges, that cannot be explained solely by the simple "microepidemic" assumption.
In order to gain new insights into bacterial evolution by simulation on larger scales, we will discuss in the talk an extension of the above simple evolutionary model, by incorporating the underlying host contact network. Moreover, by implementing an efficient and scalable simulation engine, we are able to simulate large populations over real contact networks. An added value of such model is that it allows the study of the effect of sampling on the perceived bacterial population structure as inferred by traditional methods of phylogenetic inference . The results are providing novel information that can be used in surveillance of infectious diseases and outbreak investigation and control.