A Powerful Approach to Estimating Annotation-Stratified Genetic Covariance via GWAS Summary Statistics

Abstract

Despite the success of large-scale genome-wide association studies (GWASs) on complex traits, our understanding of their genetic architecture is far from complete. Jointly modeling multiple traits' genetic profiles has provided insights into the shared genetic basis of many complex traits. However, large-scale inference sets a high bar for both statistical power and biological interpretability. Here we introduce a principled framework to estimate annotation-stratified genetic covariance between traits using GWAS summary statistics. Through theoretical and numerical analyses, we demonstrate that our method provides accurate covariance estimates, thereby enabling researchers to dissect both the shared and distinct genetic architecture across traits to better understand their etiologies. Among 50 complex traits with publicly accessible GWAS summary statistics (N≈ 4.5 million), we identified more than 170 pairs with statistically significant genetic covariance. In particular, we found strong genetic covariance between late-onset Alzheimer disease (LOAD) and amyotrophic lateral sclerosis (ALS), two major neurodegenerative diseases, in single-nucleotide polymorphisms (SNPs) with high minor allele frequencies and in SNPs located in the predicted functional genome. Joint analysis of LOAD, ALS, and other traits highlights LOAD's correlation with cognitive traits and hints at an autoimmune component for ALS.

Detecting Local Genetic Correlations with Scan Statistics

“…Compared to the local genetic correlation estimation method in the literature 12 , we do not assume genetic effects to be fixed. Instead, our framework is a direct generalization of the model developed for global genetic correlation estimation 10,11 . Under the alternative hypothesis, we denote the non-overlapping genetic regions that contribute to multiple traits to be ( , … , j and the union set as ℛ =∪ mn( j m such that ] Y [ , ] = 1 if and only if ∈ ℛ.…”

“…Simulations were based on the genotype data from the WTCCC cohort. We adopted the same quality control procedure as previously described 11 and only included SNPs on chromosome 1 in the analysis. After quality control, 15,918 individuals and 20,211 SNPs remained in the dataset.…”

“…Recently, statistical methods that jointly model multiple phenotypes have quickly gained popularity in human genetics research [1][2][3] . Leveraging pervasive pleiotropy in the human genome, these methods enhanced the statistical power to identify genetic associations 1,[4][5][6][7] , improved the accuracy of genetic risk prediction 8,9 , revealed novel genetic sharing across diverse phenotypes [10][11][12] , and provided great insights into the genetic basis of a variety of diseases and traits 13,14 .…”

“…These methods effectively utilize genome-wide genetic data, including SNPs that do not reach statistical significance in GWAS, to quantify the overall genetic sharing between two traits. In addition, recent methodological advances have enabled estimation of genetic correlation with GWAS summary statistics 10,11,22 , making these approaches widely applicable to a large number of complex phenotypes. With these advances, genetic correlation analysis has become a routine procedure in post-GWAS analysis and was implemented in almost all large-scale GWASs published in the past few years.…”

A meta‐analysis approach with filtering for identifying gene‐level gene–environment interactions

“…We randomly selected 243 genes with 5–15 common variants (MAF ≥ 5%) in LD, which provided about 2,000 variants in total. To simulate a binary response variable affected by main genetic effects only, following Lu et al [2017], we calculated the sum of minor allele count for each individual across all 2,000 variants, and then divided the samples into cases or controls based on the sums where nearly half of the samples were cases. We repeated the simulation 100 times and averaged the results for different sizes of genes.…”