Analyzing Bacterial Conjugation with Graphical Models: A Model Comparison Approach

Abstract

Conjugation is a mechanism for horizontal gene transfer that allows microbes to share genetic material with nearby cells. It plays an important role in the spread of antibiotic resistance in bacteria and is used as a tool for genetic engineering. Understanding which factors affect conjugation frequency is an ongoing challenge due to the stochastic nature of cell-cell interactions. In this thesis, we present a proof of concept of a model comparison approach for analyzing experimental data of bacterial conjugation. We develop a Bayesian network structure to model the interactions within a single experimental trial. We model different versions of biological mechanisms by assigning different conditional probability distributions to those structures. Identifying distributions that predict events consistent with the experimental results provides insight into the mechanisms governing conjugation. We compare 12 model variations for each of 6 experimental trials. Our results suggest that individual cell features and contact quality both impact the likelihood of conjugation. We also provide insight into the length of the delays involved in conjugation. These results are consistent when compared across multiple trials and metrics.