TL;DR
This study leverages multi-trait genome-wide association studies (GWAS) to identify novel genetic loci associated with optic nerve head (ONH) parameters, which are crucial for understanding glaucoma. By jointly analyzing related traits, the researchers discovered two new genetic variants that replicate across different ancestries, highlighting the power of this integrated approach for uncovering pleiotropic effects.
What Problem Does This Paper Address?
Glaucoma, a leading cause of irreversible blindness, is strongly linked to changes in the optic nerve head (ONH) morphology, including cup area, disc area, and vertical cup-disc ratio, as well as intraocular pressure (IOP). These traits are known to be highly heritable and genetically correlated, suggesting that a single genetic locus might influence multiple of these traits (pleiotropy). Traditional single-trait GWAS might miss these pleiotropic variants due to limited statistical power. This paper addresses the challenge of identifying these pleiotropic variants by employing a multi-trait GWAS approach, aiming to increase the power to detect genetic associations and uncover novel loci that might be missed by analyzing each trait in isolation.
Methods
The researchers employed two primary multi-trait analysis methods: CPASSOC and MTAG. Both methods utilize aggregated GWAS summary statistics.
- CPASSOC: This program performs a meta-analysis by considering the genetic correlation between traits. It offers two modes: SHom (assuming homogeneous genetic effects across traits) and SHet (allowing for heterogeneous effects).
- MTAG (Multi-Trait Analysis of GWAS): This method enhances the power of single-trait analyses by incorporating GWAS findings from correlated traits. It essentially boosts the signal for individual traits by leveraging information from related ones.
The study also involved:
- Meta-analysis: Combining results from multiple European ancestry cohorts (discovery) and Asian ancestry cohorts (replication).
- Imputation: Using the Haplotype Reference Consortium (HRC) panel for European cohorts and the 1000 Genomes panel for Asian cohorts to impute genotypes.
- Replication: Validating identified variants in independent Asian cohorts.
- In silico analysis: Investigating the functional and regulatory potential of the identified variants using bioinformatics tools, including ENCODE and Roadmap Epigenomics data for chromatin states and transcription factor binding sites, as well as eQTL (expression quantitative trait loci) data.
- Pathway analysis: Determining the biological functions and pathways associated with the identified genes.
- POAG validation: Testing the identified variants in independent primary open-angle glaucoma (POAG) case-control studies to assess their translational potential.
Datasets
The study utilized data from:
- European Ancestry Discovery Cohorts: 12 cohorts from the International Glaucoma Genetic Consortium (IGGC), imputed to the Haplotype Reference Consortium (HRC) release 1 panel. This dataset comprised approximately 31,269 individuals for IOP and up to 26,675 for ONH traits.
- Asian Ancestry Replication Cohorts: 5 cohorts from the IGGC, imputed to the 1000 Genomes panel.
- POAG Case-Control Studies: Three independent POAG studies (NEIGHBOR, Southampton, and UK Biobank Eye and Vision Consortium) for validation.
No new benchmark datasets were introduced; instead, the study leveraged existing large-scale genetic association data.
Key Insights and Ideas
- The power of multi-trait analysis: The study strongly demonstrates that jointly analyzing genetically correlated traits significantly increases the power to detect novel genetic variants, especially those with pleiotropic effects. This is a crucial advancement for dissecting complex genetic architectures of related phenotypes.
- Pleiotropy is key for ONH traits: The high genetic correlation observed between ONH traits (cup area, disc area, vertical cup-disc ratio) makes them ideal candidates for multi-trait GWAS, as it suggests shared genetic underpinnings.
- Novel loci identified: The multi-trait approach successfully identified two novel genetic loci (rs11158547 near PPP1R36-PLEKHG3 and rs1028727 near SERPINE3) associated with ONH morphology that were not found in single-trait analyses.
- Replication across ancestries: The identified novel variants replicated in independent Asian cohorts, strengthening their validity and suggesting broad applicability.
- Functional relevance: In silico analyses suggest that these novel variants are located in or near regulatory elements, indicating potential functional roles in gene expression and, consequently, ONH development and function.
- Limited direct link to POAG: While the identified variants are associated with ONH morphology, their direct association with POAG was not consistently replicated, likely due to high heterogeneity in POAG studies. This highlights the complexity of translating ONH findings directly to disease risk.
Key Results
- Two novel loci identified: rs11158547 (near PPP1R36-PLEKHG3) and rs1028727 (near SERPINE3) were identified at genome-wide significance through multi-trait analysis.
- Replication: These variants replicated in independent Asian cohorts.
- Consistency across methods: Findings were consistent when using both CPASSOC and MTAG multi-trait analysis approaches.
- Functional annotations: Variants in LD with the novel loci showed overlap with active chromatin states and potential regulatory elements, suggesting a role in gene regulation.
- POAG association: rs11158547 showed a Bonferroni-significant association with POAG risk in a fixed-effect meta-analysis, but this was not confirmed by a random-effects meta-analysis due to heterogeneity.
Limitations and Future Work
- Imputation panel differences: The use of different imputation panels (HRC for European, 1000 Genomes for Asian) for discovery and replication might introduce bias and hinder replication.
- Heterogeneity in POAG studies: The lack of consistent replication for the identified variants in POAG case-control studies is attributed to high heterogeneity in these cohorts, suggesting the need for larger and more homogeneous datasets.
- Translational gap: While the identified variants are strongly associated with ONH morphology, their direct clinical impact on POAG risk requires further investigation.
- Future work: The authors suggest further research, including exome sequencing and functional studies, to precisely define the roles of PPP1R36-PLEKHG3 and SERPINE3 in POAG pathogenesis. They also emphasize the need for larger and more homogeneous POAG datasets to confirm the translational potential of these findings. The study underscores the value of multi-trait analysis for future genetic studies of complex traits.
Summary updated