Genome-scale analysis of adaptation: a functional system approach

HERNÁN DOPAZO

Porto, Portugal

Conferencia; 12th Portugaliae Genetica. From genomes to organisms, an evolutionary perspective; 2009

IPATIMUP institute

Genome­Scale Analysis of Adaptation: A Functional System Approach Hernán Dopazo Comparative Genomics Unit. Bioinformatics and Genomics Department Centro de Investigación Príncipe Felipe Valencia. Spain The analysis of adaptation in large­scale studies or in complete genomes relies on methods and concepts originally conceived for the study of single genes. The current paradigm for large scale analyses of adaptation typically consists on the successive application of a test to find statistically significant deviations from neutrality (that is, significant increase above a threshold value ω=dN/dS = 1) to all the genes studied, followed by correction for multiple testing. In order to understand the functional roles under positive selection a conventional functional enrichment test, which ascertains the overabundance of gene ontology (GO) or other functional annotations, was applied to the resulting list of positively selected genes. With variations in the methods chosen to test for positive selection and/or to search for functional enrichment, the threshold­based approach described was applied in different comparative genomic studies with results below the expectations. The enriched functional categories found hardly ever reached statistical significance after correction for multiple testing. Moreover, this approach failed to detect adaptive differences between species. We describe an extensive multispecies genomic comparison in which we have applied a novel gene­set based functional analysis that has unveiled over one order of magnitude (on functional modules) more than it was previously thought1­4. The main contribution is that we have directly tested positive selection over gene modules (defined as gene ontology terms and KEGG pathways) instead of testing firstly genes independently and then looking for over­ representation of functional terms. The application of gene­set based methods has been successful in other areas, such as transcriptomics, but these have never been applied in evolutionary biology. Gene­set based methods introduce systems biology concepts in the sense that the entity tested is not the gene but a higher order unit, the module, that accounts for the functionality of the cell better that the genes in isolation. Here we demonstrate the enormous increase in testing power that the use of such concepts has in evolutionary biology. References 1. Arbiza, L., Dopazo, J. & Dopazo, H. Positive selection, relaxation, and acceleration in the evolution of the human and chimp genome. PLoS Comput Biol 2, e38 (2006). 2. Bakewell, M.A., Shi, P. & Zhang, J. More genes underwent positive selection in chimpanzee evolution than in human evolution. Proc Natl Acad Sci U S A 104, 7489­94 (2007). 3. Clark, A.G. et al. Inferring nonneutral evolution from human­chimp­mouse orthologous gene trios. Science 302, 1960­3 (2003). 4. Nielsen, R. et al. A scan for positively selected genes in the genomes of humans and chimpanzees. PLoS Biol 3, e170 (2005)