Using gene genealogies to detect rare variants associated with complex traits

Area of Science:

• Population genetics
• Statistical genetics
• Genomics

Background:

• Deleterious genetic variants are typically rare and recently introduced according to population genetic theory.
• Ancestral tree-based methods are established for common genetic variants but their utility for rare variants is less explored.
• Gene genealogies illustrate haplotype relationships, offering a framework for genetic association studies.

Purpose of the Study:

• To investigate if gene genealogies can enhance the detection of genomic regions with multiple rare causal variants.
• To develop and evaluate tree-based statistics for identifying associations with rare genetic variation.
• To determine if ancestral tree-based approaches offer advantages over non-tree-based methods for rare variant discovery.

Main Methods:

• Simulated genetic data under various demographic models.
• Developed and compared multiple tree-based statistics using known gene genealogies.
• Assessed the ability of statistics to detect clustering indicative of rare causal variants.

Main Results:

• A novel statistic, utilizing scaled distance between tree tips, demonstrated superior performance.
• This novel statistic outperformed existing tree-based methods in simulations.
• The ancestral tree-based approach showed potential for rare variant discovery, especially under mild population growth.

Conclusions:

• Gene genealogies offer a valuable framework for rare variant association studies.
• The novel tree-based statistic is a promising tool for identifying genomic regions with rare causal variants.
• Ancestral tree-based methods have potential for advancing rare variant discovery in genetic research.