Single nucleotide polymorphisms associated with rat expressed sequences

Abstract

Single nucleotide polymorphisms (SNPs) are the most common source of genetic variation in populations and are thus most likely to account for the majority of phenotypic and behavioral differences between individuals or strains. Although the rat is extensively studied for the latter, data on naturally occurring polymorphisms are mostly lacking. We have used publicly available sequences consisting of whole-genome shotgun (WGS), expressed sequence tag (EST), and mRNA data as a source for the in silico identification of SNPs in gene-coding regions and have identified a large collection of 33,305 high-quality candidate SNPs. Experimental verification of 471 candidate SNPs using a limited set of rat isolates revealed a confirmation rate of approximately 50%. Although the majority of SNPs were identified between Sprague-Dawley (EST data) and Brown Norway (WGS data) strains, we found that 66% of the verified variations are common among different rat strains. All SNPs were extensively annotated, including chromosomal and genetic map information, and nonsynonymous SNPs were analyzed by SIFT and PolyPhen prediction programs for their potential deleterious effect on protein function. Interestingly, we retrieved three SNPs from the database that result in the introduction of a premature stop codon and that could be confirmed experimentally. Two of these "in silico-identified knockouts" reside in interesting QTL regions. Data are publicly available via a Web interface (http://cascad.niob.knaw.nl), allowing simple and advanced search queries.

Figure 1

Distribution of candidate SNPs on the rat linkage group map. The density (per 1-Mb window) of synonymous (orange) and nonsynonymous (red) candidate SNPs is plotted against gene density (green) as annotated in rat genome build 3.1 (February 2004). Loci with confirmed polymorphisms are indicated by closed blue boxes.

Figure 2

Phylogenetic tree of 18 rat individuals based on a set of 82 polymorphic loci as obtained in the first verification experiment. Coefficients represent interior branch test support values for tree nodes.