The Aris-Brosou lab
Our research group works in Computational Molecular Evolution. The themes we address range from very specific theoretical aspects to applications based on real data sets (hypothesis-driven) or complete databases (both data and hypothesis-driven). Recent and current research topics include (click links below for details):
- Influenza evolution
- Viral phylodynamics
- Theoretical developments in estimation of divergence times
- Environmental genomics of the Haptophytes
- Estimation of selective pressures acting on protein-coding genes
- Mode and tempo of the diversification of gene families
Codon-based substitution models are routinely used to measure selective pressures acting on protein-coding genes. To this effect, the nonsynonymous to synonymous rate ratio (dN/dS = ω) is estimated, with ω > 1 generally indicating evidence for positive selection. Work related to codon models is divided up into three parts. First, I proposed that genes belonging to complex systems are relatively less likely to be under positive selection. This "extended complexity hypothesis," was then evaluated and supported by the analysis of 2,428 families and protein domains. Second, the impact of the parameterization of these codon models was assessed in order to evaluate their power to tease apart the roles of mutation and selection in the evolution of genomes. We showed, by means of a simulation study, that two different ways of modeling codon frequencies can have a dramatic impact on rate estimates and affect biological conclusions about genome evolution. Last, I extended a previous approach ("full Bayes") to estimate selective pressures occurring at amino-acid sites. The new procedure was found to be superior to a previous one (empirical Bayes) and was relatively robust to light misspecification of the process generating adaptive evolution. Altogether, this work suggests that both the mode and the rate of evolution of proteins are influenced by their gene ontology and by their connectivity, but that routinely used codon models are still limited descriptors of the complex dynamics of protein evolution.
Research facilities
A small computer cluster was purchased from Sun Microsystems in 2007 thanks to a CFI grant with one X2200, three X4100 and two X4600 servers clocked at 2.6 GHz with 96 GB of distributed memory (up to 32 GB on the X4600s); iMacs, PC xeon boxes and (for legacy) ultra sparc 80s are available as workstations.
