Rare coding variant analysis for human diseases across biobanks and ancestries

Area of Science:

• Genomics
• Human Genetics
• Disease Association Studies

Background:

• Large-scale sequencing data provides insights into rare genetic variations influencing human traits.
• Investigating rare coding variation is crucial for understanding phenotypic variability.

Purpose of the Study:

• To conduct a pan-ancestry analysis of rare coding variation across multiple large biobanks.
• To identify gene-disease associations for 601 diseases in a diverse population of 748,879 individuals.
• To evaluate the utility of pan-ancestry burden testing for genetic discovery in diverse datasets.

Main Methods:

• Utilized mixed-effects models for gene-based rare variant testing.
• Analyzed sequencing data from three large biobanks, including the All of Us research program.
• Included 155,236 individuals with non-European ancestry in the analysis.

Main Results:

• Identified 363 significant gene-disease associations, revealing core genes for the human disease phenome.
• Discovered potential novel associations, such as UBR3 for cardiometabolic disease and YLPM1 for psychiatric disease.
• Found concordant effect sizes for rare protein-disrupting variants across European and other genetic ancestries (βDeming = 0.7-1.0).

Conclusions:

• Pan-ancestry burden testing is an effective approach for genetic discovery in diverse populations.
• Ancestry-specific sensitivity analyses are important for comprehensive genetic studies.
• Findings support multi-ancestry and cross-biobank strategies for sequencing association studies in human disease.