A genome-wide association study for blood lipid phenotypes in the Framingham Heart Study

Using a 100K genome-wide scan, we have generated a set of putative associations for common sequence variants and lipid phenotypes. Validation of selected hypotheses in additional samples did not identify any new loci underlying variability in blood lipids. Lack of replication may be due to inadequate statistical power to detect modest quantitative trait locus effects (i.e., <1% of trait variance explained) or reduced genomic coverage of the 100K array. GWAS in FHS using a denser genome-wide genotyping platform and a better-powered replication strategy may identify novel loci underlying blood lipids.