To be able to predict evolution, we must know something about the mutations that contribute to adaptive evolution. Having the ability to predict adaptive evolution is important when faced with threats like antibiotic resistant ‘super bugs’ or evolving zoonotic pathogens. Most of what we know about the characteristics of beneficial mutations comes from theory involving populations that are already really well adapted to their environment. In reality, populations can be vastly less fit than theory assumes.
I study adaptive evolution using using microbes as a model system. Presently, I am working with the opportunistic pathogen Pseudomonas aeruginosa to study the emergence of resistance and return to sensitivity after short term exposure to antibiotics. I am also involved the bioinformatics analysis of projects on beta-lactam resistance (with Victoria Furio), mutational robustness (with Karl Heilbron), population size and the repeatability of resistance evolution (with Tom Vogwill) and phage/bacteria co-evolution (with Alex Betts).
In my DPhil, I used rifampicin resistant P. aeruginosa adapting to use a non-preferred carbon substrate (L-serine) as a model system. Different resistance mutations seem to confer a different ability to adapt to this substrate, which may be due to interactions between rpoB and a gene called PA2449. PA2449 is a transcriptional regulator involved in glycine and serine metabolism, as well as pyocyanin biosynthesis.
In my MSc work, I studied compensatory adaptation of Aspergillus nidulans in a complex medium. Using a maximum likelihood-based model, I found that the number of major fitness jumps during adaptive walks is independent of starting fitness—populations undergo two major changes in fitness, on average. On-going research will determine whether the number of fitness increases corresponds to the number of mutations fixed.