Deviation from baseline mutation burden provides powerful and robust rare-variants association test for complex...

Area of Science:

• Genetics
• Biostatistics
• Genomic Medicine

Background:

• Identifying rare genetic variants for complex diseases is difficult due to low statistical power in traditional case-control tests.
• Existing methods often fail to detect significant associations involving rare variants.

Purpose of the Study:

• To develop a novel and powerful rare variant association test to improve the identification of genes contributing to complex diseases.
• To enhance the detection of susceptibility genes with rare risk variants.

Main Methods:

• Proposed a new rare variant association test named Recursive Unbiased Negative-binomial REgression (RUNNER).
• RUNNER utilizes weighted recursive truncated negative-binomial regression on genomic features to predict baseline mutation burden.
• Evaluated performance through simulations and application to real case-control and case-only genetic data.

Main Results:

• RUNNER demonstrated substantially higher power compared to state-of-the-art rare variant association tests in simulations.
• The method maintained reasonable type 1 error rates across diverse populations and sample sizes.
• Applied to Hirschsprung and Alzheimer's disease data, RUNNER identified known genes missed by other methods and suggested new candidates.
• Successfully detected a known causal gene for amyotrophic lateral sclerosis in a case-only analysis.

Conclusions:

• RUNNER is a powerful and robust statistical method for identifying susceptibility genes associated with complex diseases.
• The approach effectively leverages genomic features to improve the detection of rare risk variants.
• This method offers a significant advancement in the field of genetic association studies for complex disorders.