A cross-population atlas of gene-environment interactions comprising 440,210 individuals from European and Japanese populations with replication in 539,794 individuals reveals how gene-environment interactions uncover missing heritability, affect polygenic prediction accuracy and cross-population portability, and identify sex-discordant genetic effects in lipid metabolism informing clinical trial failures.
Key Findings
Methods
A cross-population compendium of gene-environment interactions was constructed using over 440,000 individuals from European and Japanese populations.
The atlas comprised 440,210 individuals from European and Japanese populations.
Replication was performed in 539,794 individuals from diverse populations.
The study decomposed contributions from age, sex, and lifestyles to delineate the aetiology of gene-environment interactions.
Both European and Japanese populations were included to enable cross-population analysis.
Results
Gene-environment interactions were found to uncover missing heritability in complex traits.
Genome-wide analyses uncovered missing heritability connected by synergistic effects of genome and environments.
Trait-trait relationships were also revealed through these synergistic genome-environment effects.
The findings suggest that environmental modulation of genetic effects accounts for a portion of heritability not captured by standard genome-wide association studies.
Results
Gene-environment interactions systematically affected polygenic prediction accuracy and cross-population portability of polygenic scores.
The synergistic effects of genome and environments systematically affected polygenic prediction accuracy.
Cross-population portability of polygenic scores was also systematically influenced by gene-environment interactions.
These findings have implications for personalized medicine and the transferability of genetic risk scores across populations.
Results
A reverse-causality mechanism was identified in which disease-related dietary changes confounded apparent gene-environment interactions.
By decomposing contributions from age, sex, and lifestyles, the study delineated the aetiology of gene-environment interactions.
One identified mechanism was reverse-causality arising from disease-related dietary changes.
This finding highlights a methodological challenge in interpreting dietary gene-environment interactions in observational studies.
Results
Single-cell projection revealed aging-related shifts in the pathways and cell types responsible for genetic regulation.
Single-cell projection was used to map gene-environment interaction signals onto specific cell types and pathways.
An aging shift of pathways and cell types responsible for genetic regulation was identified.
This finding connects age as an environmental modifier to specific cellular mechanisms underlying genetic associations.
Sex-discordant genetic effects were identified specifically in lipid metabolism pathways.
These sex-discordant effects were identified through omics-level gene-environment analyses.
The identified sex-discordant genetic effects were found to inform clinical trial failures for genetically supported drug development.
The findings have implications for understanding why certain lipid-targeting therapies may differ in efficacy between sexes.
Conclusions
The study offers a comprehensive framework for decoding the dynamics of genetic associations across populations and environments.
The authors describe the work as a 'comprehensive gene-environment study' that 'decodes the dynamics of genetic associations.'
The study offers insights into 'complex trait biology, personalized medicine and drug development.'
Both European and Japanese populations were studied as discovery cohorts, with diverse populations used for replication, enabling cross-population generalization.