Although many genome-wide association studies (GWAS) have sought to identify genetic variants that are associated with disease, all such studies consider the two parental alleles as having identical effects. As a result, they are severely underpowered to detect effects such as imprinting that can operate differentially between the maternal and paternal alleles. We have developed novel SNP association methodologies that utilize mother/father/child trios, allowing the detection of parent of origin biases in both quantitative traits and disease association studies. We show that the use of these novel strategies that analyze disease associations separately for the maternal and paternal alleles are able to detect disease susceptibility genes that are missed using conventional GWAS approaches. In this way imprinting and other effects that can operate differentially between mother and father (e.g. maternal/fetal interactions) can be detected. In our preliminary data, we demonstrate the power of this approach to detect novel genes showing PofO effects in oral clefts. We now propose to apply this methodology to reanalyze in detail several large GWAS datasets for which SNP data from complete trios are available through The Database of Genotypes and Phenotypes (dbGAP). These studies will likely reveal novel PofO effects operating in Oral Cleft Lip/Palate (OCL/P), Attention Deficit Hyperactivity Disorder (ADHD) and Multiple Sclerosis (MS). This proposal will provide novel insights into the influence of imprinting and maternal/fetal interactions in several human diseases.
Hundreds of genome wide studies have now been performed in an attempt to identify genes underlying human disease risk. However, despite overwhelming evidence that some diseases are inherited more often from one parent than the other, most genome-wide association studies consider the two parental alleles to be equivalent, severely limiting their ability to detect these parent of origin effects. We have developed novel methods that can reliably detect parent of origin biases in disease risk, and here we propose to apply these methods to reanalyze data from published studies of three diseases.
Showing the most recent 10 out of 14 publications
