Multi-trait GWAS for diverse ancestries: Mapping the knowledge gap

Abstract

Background: Approximately 95% of samples analyzed in univariate genome-wide association studies (GWAS) are of European ancestry. This bias toward European ancestry populations in association screening also exists for other analyses and methods that are often developed and tested on European ancestry only. However, existing data in non-European populations, which are often of modest sample size, could benefit from innovative approaches as recently illustrated in the context of polygenic risk scores. Methods: Here, we extend and assess the potential limitations and gain of our multi-trait GWAS pipeline, JASS (Joint Analysis of Summary Statistics), for the analysis of non-European ancestries. To this end, we conducted the joint GWAS of 19 hematological traits and glycemic traits accross five ancestries (European (EUR), admixed American (AMR), African (AFR), East Asian (EAS), South-East Asian (SAS)). Results: We detected 367 new genome-wide significant associations in non-European populations (15 in Admixed American (AMR), 72 in African (AFR) and 280 in East Asian (EAS)). New associations detected represent 7%, 25% and 21% of associations in the AFR, AMR and EAS populations, respectively. Overall, multi-trait testing increases the replication of European associated loci in non-European ancestry by 15%. Pleiotropic effects were highly similar at significant loci across ancestries (e.g. the mean correlation between multi-trait genetic effects of EUR and EAS ancestries was 0.88). For hematological traits, strong discrepancies in multi-trait genetic effects are tied to known evolutionary divergences: the ARKC1 loci, which is adaptive to overcome the p.vivax induced malaria. Conclusions: Multi-trait GWAS can be a valuable tool to narrow the genetic knowledge gap between European and non-European populations.

Trait selection strategy in multi-trait GWAS: Boosting SNPs discoverability

“…Figure 3 presents the correlation between each feature and the number of univariate and multi-trait GWAS associated loci. As expected, the number of univariate associated loci was positively correlated with the number of traits, mean Neff, the mean h can lead to a lack of robustness of the omnibus test 10 . We checked how the condition number 𝜅 * was related to JASS gain (Fig.…”

“…Overall, this suggests that the multi-trait test can be highly complementary to the univariate test, performing better in situations where the univariate tests display low power. We noted in a recent study 10 that a high multicollinearity of the matrix underlying the null hypothesis (Σ r ) can lead to a lack of robustness of the omnibus test 10 . We checked how the condition number Σ r was related to JASS gain ( Fig.…”

“…Selecting clinically homogenous traits is the strategy most commonly used 4,8, 10,14–19 . In our previous large-scale analysis 8 , an heterogeneous set yielded the largest number of new association as compared to clinically homogenous sets.…”

“…This focus should not lead the reader to think that multi-trait GWAS is useful only on large sample studies of European ancestry. We actually recently updated the JASS pipeline to run a Multi-ancestry Multi-trait GWAS, which was able to detect 367 new association loci, despite the modest sample size of the non-European cohorts used 10 . Future work might leverage non-European existing 28 and upcoming biobanks 29 to investigate the validity of our results for non-European ancestries.…”

Genetic Inheritance Models of Non-Syndromic Cleft Lip with or without Palate: From Monogenic to Polygenic

“…One of the pitfalls of GWAS is that the result is easily confounded by different ancestries of selected samples, case-control imbalance, various sequencing methods, and linkage disequilibrium, leading to poor stability and transferability across different study groups [37][38][39]. However, with the progression of GWAS analysis tools, these confounding factors could be minimized to a lower level.…”