22924
A Comparison of Polygenic Contribution to Autism Spectrum Disorder Using Common, Rare and Copy Number Variants
Background: Current evidence suggests that the genetic contribution to autism spectrum disorder (ASD) is polygenic, consisting of small, accumulating effects of many variants. While some of the contribution may come from rare single variant mutations or rare copy number variation, the summation of modest effects among common variants is still likely to be important.
Objectives: To create polygenic risk scores (PRS) using large sample results from the Psychiatric Genomics Consortia (PGC) mega-analysis for ASD using genome-wide genotype data from the Study to Explore Early Development (SEED), a multi-site case-control study and compare the performance of these PRS to scores calculated from copy number variant burden, rare variant burden, or other combined metrics.
Methods: We have created a polygenic risk score, derived from discovery results available from the PGC for ASD, using measured genotypes in SEED to investigate the interplay of genetic and environmental risk factors that underlie disease risk. We will further create a score reflective of the burden of rare variation (using 1000 genomes imputed data) and rare copy number variation burden (using Penn CNV calls). Correlations among the three scores, their potential interactions, and their ability to classify membership into case status will be explored.
Results: The association of the common variant PRS with ASD status was significant, but appeared to be due to confounding by genetic ancestry. We will present results adjusted for ancestry as well as comparisons with rare-variant based scores.
Conclusions: This is the first effort to our knowledge to investigate multiple polygenic scores in ASD. Our results can shed light on whether rare, common and CNV burden are correlated or represent independent contributions to the risk of autism.