February 12, Wed 2013
3:30 pm, 1005 Haworth
Dr. David M. Kristensen
Computational Biology Branch, National Center for Biotechnology Information
Orthology & Bacterial Pathogenicity
In order to probe deeply into the mechanisms of microevolutionary processes such as phage-mediated transduction of virulence factors that transforms normal, healthy bacteria into pathogens, I develop tools and approaches that facilitate these types of comparative genomics studies of microbiota (bacteria, archaea, and the viruses that infect them). For example, I recently aided my collaborators in performing the first large-scale, high-resolution study of the viral diversity that is present within the human intestinal tract, which in healthy individuals, consists primarily of phages that infect the bacterial communities commonly found therein. This work was expedited by using a database of evolutionary gene families in phages to identify signature genes that act as sensitive and precise diagnostic indicators of the presence and quantitative abundance of a given type of virus in a metagenomic sample.
In another example, I will also discuss the identification of highly degraded prophage regions within the ~2500 completely sequenced bacterial and archaeal genomes currently available. Of particular interest is the domestication process whereby bacteria transform their would-be predators into molecular machinery that provides a competitive advantage over their uninfected brethren (e.g., adding virulence factors that may produce toxins, increase cell growth and/or mutation rate, or alter cell adhesion or other interactions with their environment). Despite the propensity of prophages to degrade quickly after lysogenization, a procedure involving advanced homology detection and gene density cluster is able to identify both known and novel instances of prophage domestication. Furthermore, these regions often demonstrate measurably perturbed evolutionary signatures in comparison to their bacterial hosts, even while retaining diagnostic signatures of the specific viral group from which they originated.
The fully automated, computationally efficient processes to build these databases will allow these resources to continue to provide insight into microbial genomes in the future, even as the rapid pace of genome sequencing continues unabated.