LAVA: An integrated framework for local genetic correlation analysis

Abstract

Genetic correlation (rg) analysis is commonly used to identify traits that may have a shared genetic basis. Traditionally, rg is studied on a global scale, considering only the average of the shared signal across the genome; though this approach may fail to detect scenarios where the rg is confined to particular genomic regions, or show opposing directions at different loci. Tools dedicated to local rg analysis have started to emerge, but are currently restricted to analysis of two phenotypes. For this reason, we have developed LAVA, an integrated framework for local rg analysis which, in addition to testing the standard bivariate local rg’s between two traits, can evaluate the local heritability for all traits of interest, and analyse conditional genetic relations between several traits using partial correlation or multiple regression. Applied to 20 behavioural and health phenotypes, we show considerable heterogeneity in the bivariate local rg’s across the genome, which is often masked by the global rg patterns, and demonstrate how our conditional approaches can elucidate more complex, multivariate genetic relations between traits.

Genome-wide association studies

“…LDSC 174 Assessment of genetic correlation between phenotypes using summary statistics as input; has various other functions, including partitioned SNP-based heritability and assessment of selection bias GCTA 173 Assessment of genetic correlation between phenotypes using raw genotypic data as input SumHer 264 Assessment of genetic correlation between phenotypes using summary statistics as input; has various other functions, including partitioned SNP-based heritability and assessment of selection bias superGNOVA 183 Assessment of local genetic correlations using GWAS summary statistics ρ-HESS 184 Assessment of local SNP-based heritability and genetic correlations using GWAS summary statistics LAVA 185 Assessment of local multivariate genetic correlations using GWAS summary statistics GenomicSEM 265 Assessment of multivariate genetic correlations based on GWAS summary statistics…”

“…Both linkage disequilibrium score regression and genome-wide complex trait analysis allow the estimation of genetic correlations, or the extent to which genetic variants that account for a trait are also important for another trait, provided that the effects are in the same direction. Tools such as superGNOVA 183 , ρ-HESS 184 and LAVA 185 from a recent preprint article allow the estimation of local correlations, determining which specific genomic regions exert genetic effects on the correlated phenotypes in the same or opposing directions. Genetic correlations should be interpreted in the context of SNP-based heritabilities; for example, if these are low for the respective phenotypes, genetic correlation is not expected to play a major part in explaining why two traits correlate at the phenotypic level.…”

On the interpretation of transcriptome-wide association studies

“…To evaluate the severity of the issues that arise when interpreting TWAS as testing genetically mediated relationships between gene expression and phenotype, we performed extensive simulations and applied TWAS to real data. To serve as a reference, we used the local genetic correlation analysis in LAVA 8 , which directly tests the true genetic covariance cov( , ). To simplify comparison, we implemented TWAS analysis inside the LAVA framework, using the same preprocessing and test statistic as for the local genetic correlation analysis, ensuring that the only difference between the two analyses is the null model being evaluated (see Methods -LAVA implementation of TWAS).…”

“…has a non-central Wishart sampling distribution, which is used to obtain p-values to test cov( , ) = 0 using a simulation procedure (see Werme et al (2021) for details).…”

“…The LAVA implementation of local genetic correlation analysis has been described in detail in Werme et al (2021) 8 . In brief, LAVA uses summary statistics and a reference genotype sample to fit equations…”