Detecting rare variants for complex traits using family and unrelated data

Large genome-wide association studies (GWAS) have been performed to detect common genetic variants involved in common diseases, but most of the variants found this way account for only a small portion of the trait variance. Furthermore, candidate gene-based resequencing suggests that many rare genetic variants contribute to the trait variance of common diseases. Here we propose two designs, sibpair and unrelated-case designs, to detect rare genetic variants in either a candidate gene-based or genome-wide association analysis. First we show that we can detect and classify together rare risk haplotypes using a relatively small sample with either of these designs, and then have increased power to test association in a larger case-control sample. This method can also be applied to resequencing data. Next we apply the method to the Wellcome Trust Case Control Consortium (WTCCC) coronary artery disease (CAD) and hypertension (HT) data, the latter being the only trait for which no genome-wide association evidence was reported in the original WTCCC study, and identify one interesting gene associated with HT and four associated with CAD at a genome-wide significance level of 5%. These results suggest that searching for rare genetic variants is feasible and can be fruitful in current GWAS, candidate gene studies or resequencing studies.