Adaptations are central to the study of evolution. Thus it is surprising that we know so little about the molecular basis of adaptive evolution. The goal of my research is to identify, clone, and characterize the evolution of genes underlying natural adaptations in order to determine the types of genes involved, how many and what types of genetic changes occurred, and the evolutionary history of these changes. These data will address key questions. For example, do adaptations involve many genetic changes or only a few? How important are regulatory versu! s amino acid changes in adaptation? How often are “new” gene s involved in adaptations? Are most adaptive alleles new mutations or pre-existing alleles segregating at low to moderate frequency within a species? Clearly, a deeper understanding of how genes change during adaptation will give insight into the potential and limits of adaptive evolution.
Spatial patterns of genomic features
I am also using genomic data to address important evolutionary questions. Recently, I used D. melanogaster genome sequence data to estimate genome-wide levels of gene clustering and to contrast the amount of clustering among genes with similar motifs to the levels of clustering in general. All chromosomes, except the fourth, showed substantial levels of gene clusteri! ng. Although not more clustered than the average pair of adjacent genes, genes with the same primary motif occur adjacent to one another more often than expected by chance. These results may mean that these small local groups of genes share regulatory elements and evolutionary histories.
Detecting natural selection in DNA sequence data
Molecular evolutionists have long sought to determine which changes within the protein coding and regulatory regions of a gene were shap! ed by natural selection. If an adaptive substitution has occurred in the recent past, there should be a paucity of DNA polymorphism surrounding the site under selection. Taking advantage of this fact, Andrew Kern and I have developed a permutation approach for detecting selected sites using polarized DNA polymorphism and divergence data. This method is especially useful for detecting the effects of weak selected forces across several loci. We used this approach to analyze a large DNA polymorphism and divergence data set of D. simulans genes. Surprisingly, although replacement fixations do not on average appear to be driven by selection, preferred codons – those codons that use the most abundant tRNA – have on average been fixed by selection. We plan to apply this method to additional data sets and to look at spatial patterns of nucleotide fixation within and around genes. For instance, one could see if there is a bias in the types of polymorphism (sy! nonymous vs. non-synonymous) nearest to a type of fixation. This would give insight into the dynamics of the fixation process and its impacts on adjoining variation.