|
Abstract:
|
A central goal in the biological sciences is the identification of links between quantitative traits and locations along the genome (single nucleotide polymorphisms, or SNPs) to better understand variation in complex traits. Although quantitative trait mapping methods have identified some such links, progress has been stinted by an inability of these techniques to consider complex, but biologically realistic, scenarios such as traits affected by multiple SNPs or interactions among SNPs. Some techniques that more fully use information in existing data do so by using a phylogeny to represent evolutionary relationships among randomly sampled individuals at each SNP. Although such methods have shown promise in association mapping, these methods have just begun to consider complex data sets. Moreover, approaches that allow for multiple SNPs to affect a single trait have been limited to regression-based approaches that are computationally feasible but fail to consider shared evolutionary histories. Here, a Bayesian phylogenetic model that allows for multiple SNP effects is proposed, and parameter estimates from existing and proposed methods are compared.
|
