Genome-wide association studies (GWAS) have been used to idnetify SNPs associated with complex traits. SNP arrays that assay up to 2.5 million polymorphisms have been used and these data have often been augmented by imputation of non-genotyped variants using information from the HapMap or 1000 Genomes Project. The recent decrease in Exome-sequencing cost has provided for applications in this area. Pasaniuc et al. show that, genome-wide SNP genotypes can be inferred at a mean r2 of 0.71 using off-target data (0.24Ã— average coverage, at an effective sequencing cost of $10â€“100 per sample) in a whole-exome study of 909 samples. The association statistics obtained using extremely low-coverage sequencing data, combined with genotype calling using 1000 Genomes, attain similar P values at known associated variants as data from genotyping arrays, without increasing false positives.The authors conclude that the effective sample size per unit cost of this approach is several times greater than for the standard GWAS study design using SNP arrays and will further increase as sequencing costs drop.
Extremely low-coverage sequencing and imputation increases power for genome-wide association studies. Pasaniuc et al. Nat Genet. 2012 May 20. PMID: 22610117
posted by Yannis Trakadis MD, MSc