SNPs in Multi-species Conserved Sequences (MCS) as useful markers in association studies: a practical approach

Abstract

Background: Although genes play a key role in many complex diseases, the specific genes involved in most complex diseases remain largely unidentified. Their discovery will hinge on the identification of key sequence variants that are conclusively associated with disease. While much attention has been focused on variants in protein-coding DNA, variants in noncoding regions may also play many important roles in complex disease by altering gene regulation. Since the vast majority of noncoding genomic sequence is of unknown function, this increases the challenge of identifying "functional" variants that cause disease. However, evolutionary conservation can be used as a guide to indicate regions of noncoding or coding DNA that are likely to have biological function, and thus may be more likely to harbor SNP variants with functional consequences. To help bias marker selection in favor of such variants, we devised a process that prioritizes annotated SNPs for genotyping studies based on their location within Multi-species Conserved Sequences (MCSs) and used this process to select SNPs in a region of linkage to a complex disease. This allowed us to evaluate the utility of the chosen SNPs for further association studies. Previously, a region of chromosome 1q43 was linked to Multiple Sclerosis (MS) in a genome-wide screen. We chose annotated SNPs in the region based on location within MCSs (termed MCS-SNPs). We then obtained genotypes for 478 MCS-SNPs in 989 individuals from MS families.

Results: Analysis of our MCS-SNP genotypes from the 1q43 region and comparison to HapMap data confirmed that annotated SNPs in MCS regions are frequently polymorphic and show subtle signatures of selective pressure, consistent with previous reports of genome-wide variation in conserved regions. We also present an online tool that allows MCS data to be directly exported to the UCSC genome browser so that MCS-SNPs can be easily identified within genomic regions of interest.

Conclusion: Our results showed that MCS can easily be used to prioritize markers for follow-up and candidate gene association studies. We believe that this novel approach demonstrates a paradigm for expediting the search for genes contributing to complex diseases.

Figure 1

Scheme for identifying MCS-SNPs.

Figure 2

UCSC genome browser image showing position of MCS-SNPs in the vicinity of the EXO1 gene. After obtaining the subset of SNPs within MCS, the Table Browser function allows MCS-SNPs to be downloaded in table format or visualized as a custom track on the browser as shown here. Note several MCS-SNPs are in EXO1 5' flanking region, and in introns (e.g. rs4149896, rs2526698) as well as those in exons.