LD Score Regression Distinguishes Confounding from Polygenicity in Genome-Wide Association Studies

Abstract

Both polygenicity1,2 (i.e. many small genetic effects) and confounding biases, such as cryptic relatedness and population stratification3, can yield inflated distributions of test statistics in genome-wide association studies (GWAS). However, current methods cannot distinguish between inflation from bias and true signal from polygenicity. We have developed an approach that quantifies the contributions of each by examining the relationship between test statistics and linkage disequilibrium (LD). We term this approach LD Score regression. LD Score regression provides an upper bound on the contribution of confounding bias to the observed inflation in test statistics and can be used to estimate a more powerful correction factor than genomic control4-14. We find strong evidence that polygenicity accounts for the majority of test statistic inflation in many GWAS of large sample size.