Hi-C informed kernel association test: integrating 3-dimensional genome structure into variant-set association for...

Area of Science:

• Genomics
• Bioinformatics
• Statistical Genetics

Background:

• Variant-set association analysis is crucial for whole genome sequence (WGS) data, particularly for rare variants.
• Three-dimensional (3D) genome architecture significantly influences gene transcription and regulatory processes.
• Existing methods primarily focus on gene-centric association tests, limiting whole-genome applications.

Purpose of the Study:

• To extend 3D genome-guided association testing from gene-centric to gene-agnostic, whole-genome analysis.
• To develop a novel Hi-C informed kernel association test for improved rare variant detection.
• To integrate 3D genome architecture into genetic similarity kernels for enhanced statistical power.

Main Methods:

• Developed a principled procedure to convert Hi-C contact confidence into borrowing weights.
• Integrated these weights into genetic similarity kernels for association testing.
• Introduced a controlling parameter for adaptive information borrowing from interacting loci.

Main Results:

• The Hi-C informed kernel association test demonstrated improved performance in simulations.
• The method showed advantages in detecting rare-variant sets using WGS data.
• Successfully applied the test to the ARIC study data from the Trans-Omics for Precision Medicine (TOPMed) program.

Conclusions:

• The Hi-C informed kernel association test offers a powerful gene-agnostic approach for WGS data analysis.
• Integrating 3D genome architecture enhances the detection of rare variant sets.
• This method holds promise for advancing genetic studies and precision medicine.