Detection of natural selection in candidate genes: approximation to the bases of human adaptation

Modern human genetic diversity is the result of demographic history, and selective effects that have acted to adapt different populations to their environments. The availability of genomewide sets of SNPs such as those analyzed by the Perlegen company or the International HapMap project, has provided SNP allele frequencies in different populations, and has allowed genomewide scans for the detection of natural selection in humans. Several published studies present lists of genes, which according to their pattern of population differentiation or unusual extension of linkage disequilibrium, are suggested as candidates to have been subjected to postive selection in one or more human populations; however, no inference on the biological or functional basis for adaptation has been attempted. Evidence for selection in such genome regions may be due to type I error, and, in any case, it has been obtained through partial genomic information and without a clear possibility to ascertain the functional changes underlying the assumed adaptation. From these candidate genes, we propose to filter around 20 gene regions and to resequence them in 60 individuals (20 Europeans, 20 Asians and 20 Yorubas). We will prioritize the regions coding for functions that are related to characters that may have been selected during human evolution such as those involved in nutrition, metabolism, disease or interaction with pathogens, among others. In those pre-selected regions we will search for functional genetic variants (non-synonymous substitutions, etc), and preference will be given to those regions in which the genotyped SNPs show an allele frequency pattern compatible with a selective sweep. Resequencing will allow us to determine the complete unbiased allele frequency spectrum of all the genetic variants of these regions and therefore, we will be able to apply the classical tests of neutrality (such as Tajima’s D), allowing to accept or refuse neutral evolution. Also, in those regions in which the imprint of selection is confirmed we will design functional or expression assays in order to determine the molecular bases of the functional adaptation that have been subjected to positive selection.