Abstract |
An interesting research topic in population genetics is the detection of loci that are targets
of positive selection using Single Nucleotide Polymorphism (SNP) data. Many tests have
been developed for the identification of genomic signatures of positive selection based on the
skew of the distribution of SNP frequencies (Site Frequency Spectrum or SFS) that is caused
by selection. There are several methods to detect strong positive selection or in other words,
a selective sweep event. Current methods for detecting selective sweeps based on SFS do
not take into consideration the linkage disequilibrium in combination with the skew in the
allele frequencies in SFS. In this study we describe a model that combines both genomic
signatures of positive selection, that is, LD and the skewed patterns of allele frequencies. By
extensive simulations we show that is possible to detect a selective sweep by using the joint
distribution of the allele frequencies of pairs of polymorphic sites (joint-SFS). Furthermore,
we show that using pairs of SNPs, instead of independent sites, results in higher efficiency to
detect selective sweeps. Furthermore, we try to find a mathematical formula that calculates
the probability of observing a pair of Single Nucleotide Polymorphisms (SNPs) at a certain
distance from a positively selected site, using a diffusion approximation.
|