The flashfm approach for fine-mapping multiple quantitative traits

N Hernández,I Barroso,C Wallace,C Wallace,J L Asimit,M Sandhu,P Newcombe,P Newcombe,J Soenksen,J Soenksen

doi:10.1038/s41467-021-26364-y

Abstract

Joint fine-mapping that leverages information between quantitative traits could improve accuracy and resolution over single-trait fine-mapping. Using summary statistics, flashfm (flexible and shared information fine-mapping) fine-maps signals for multiple traits, allowing for missing trait measurements and use of related individuals. In a Bayesian framework, prior model probabilities are formulated to favour model combinations that share causal variants to capitalise on information between traits. Simulation studies demonstrate that both approaches produce broadly equivalent results when traits have no shared causal variants. When traits share at least one causal variant, flashfm reduces the number of potential causal variants by 30% compared with single-trait fine-mapping. In a Ugandan cohort with 33 cardiometabolic traits, flashfm gave a 20% reduction in the total number of potential causal variants from single-trait fine-mapping. Here we show flashfm is computationally efficient and can easily be deployed across publicly available summary statistics for signals in up to six traits.

Highlights

Joint fine-mapping that leverages information between quantitative traits could improve accuracy and resolution over single-trait fine-mapping
Flashfm addresses different statistical challenges than MFM, as we have multiple quantitative traits measured on the same individuals, so we need to account for correlation between the traits
As the traits are measured on the same individuals, it requires the trait covariance matrix, which may be calculated from an in-sample study or approximated from the Genome-wide association studies (GWAS) and trait summary statistics[23]

Summary

Introduction

Joint fine-mapping that leverages information between quantitative traits could improve accuracy and resolution over single-trait fine-mapping. Fine-mapping multiple traits could give an improvement in fine-mapping accuracy and resolution, analogous to the power increase for multi-trait GWAS, but this is computationally challenging due to the many possible combinations of models (allowing multiple causal variants) between traits, which is not an issue for multi-trait GWAS that involve testing only one SNP for association with multiple traits. For this reason, few methods exist for fine-mapping with more than two traits

Methods

Results

Conclusion