Disentangling the Effects of Colocalizing Genomic Annotations to Functionally Prioritize Non-coding Variants within...

Area of Science:

• Genomics
• Statistical Genetics
• Bioinformatics

Background:

• Fine-mapping disease loci requires distinguishing causal variants from trait-associated ones.
• Colocalization of non-coding functional annotations complicates the identification of causal variants.
• Existing enrichment methods can be biased by local genomic structure.

Purpose of the Study:

• To develop a statistical method, GoShifter, for assessing the ability of enriched annotations to prioritize causal variation.
• To address limitations of existing methods in handling colocalizing annotations and local genomic biases.
• To improve the fine-mapping of causal variants in complex diseases.

Main Methods:

• Developed Genomic Annotation Shifter (GoShifter), a statistical approach using local annotation shifting.
• Defined a null distribution for annotations overlapping alleles by shifting annotations locally.
• Compared GoShifter to SNP-matching enrichment methods, assessing sensitivity to local genomic structure.

Main Results:

• GoShifter confirmed that expression quantitative trait loci variants drive gene expression via DNase-I hypersensitive sites (DHSs) near transcription start sites and 3' UTR regulation.
• 15%-36% of trait-associated loci map to DHSs independently of other annotations.
• Identified specific patterns for breast cancer, rheumatoid arthritis, and height trait loci, prioritizing causal variants near histone mark summits and in embryonic stem cell DHSs.

Conclusions:

• GoShifter effectively prioritizes causal variation at specific loci by overcoming annotation colocalization and local genomic biases.
• The method enhances the ability to fine-map causal variants, contributing to a better understanding of disease genetics.
• GoShifter's local shifting approach offers a more robust alternative to traditional enrichment methods.