Title
Fine Mapping the Genetic Basis of CRP Regulation in African Americans: a Bayesian Approach
UMMS Affiliation
Harvard Medical School / Broad Institute; London School of Hygiene and Tropical Medicine
Faculty Advisor
Daly, Mark J.; Whittaker, John C.
Contributor(s)
Rhodes, Benjamin; Morris, David L.; Aubin, Cristin; Kelly, Jeremiah F.; Evans, Dennis A.; Mendes de Leon, Carlos F.; Oksenburg, Jorge; de Jager, Philip L.; Vyse, Tim
Date
May 2008
Document Type
Abstract
Subjects
C-Reactive Protein; African Americans; Cardiovascular Diseases; Bayes Theorem; Quantitative Trait Loci
Abstract
Basal levels of C-reactive protein (CRP), have been associated with disease, particularly future cardiovascular disease. Twin studies estimate 50% CRP heritability, so the identification of genetic variants influencing CRP expression is important. Existing Caucasian studies have identified numerous cis-acting variants but leave significant ambiguity over the key functional polymorphisms. We addressed this by typing a dense map of CRP single nucleotide polymorphisms (SNPs), and quantifying serum CRP in 594 unrelated African Americans. We used Bayesian model choice analysis to select the combination of SNPs best explaining basal CRP and found strong support for triallelic rs3091244 alone, with the T allele acting in an additive manner (Bayes factor >100 vs. null model), with additional support for a model incorporating both rs3091244 and rs12728740. Admixture analysis suggested SNP rs12728740 segregated with haplotypes predicted to be of recent Caucasian origin. Using a cladistic approach (Treescan) we confirmed the importance of rs3091244(T) by demonstrating a significant partition of haplotype effect based on the rs3091244(C/T) mutation (F=8.91, P=0.006). We argue that weaker linkage disequilibrium across the African American CRP locus compared with Caucasians has allowed us to establish an unambiguous functional role for rs3091244(T), while also recognising the potential for additional functional mutations present in the Caucasian genome.
