New computational methods for metagenomics

Microbial life is responsible for an immense number of medical and environmental processes. The role of microbes in infectious disease is well known, but bacteria also provide essential services to us, including extracting important nutrients from our food. In the environment, microbes are essential in the turnover of nutrients, breaking down and building up important molecules in a cycle that is global in scope.By sequencing microbial genomes, we can learn a great deal about the roles they play and the myriad complex ways in which they interact. But the sheer amount of data is overwhelming, and the rate of increase in computer speed is being overtaken by the flood of sequence data. We need new and better computational methods to be able to take a random sampling of DNA from the environment and assess its function and its history. Accurate analysis will allow us to highlight the key genes and organisms from a given environment, opening up new possibilities for control and harnessing the potential of these microbial populations. We also stand to learn a great deal about the evolutionary history of these organisms and populations, shedding light on four billion years of microbial evolution.The goal of my research is to build computational tools that are fast, accurate, and statistically sound, which can meet the challenges of emerging genome data sets. Important parts of this work include applications for the assignment of functions to newly discovered genes, new methods to track the evolutionary histories of genes and whole genomes, and simulation methods to gauge the accuracy of our conclusions. An important unifying theme is the use of evolutionary history to assess and understand microbial diversity.