Ancestral recombination graphs under non-random ascertainment, with applications to gene mapping

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Genetic linkage analysis is crucial for identifying disease susceptibility genes.
• Existing methods often struggle with complex inheritance patterns and unphased genotype data.

Purpose of the Study:

• To develop a novel gene mapping algorithm for identifying disease susceptibility genes.
• To model shared ancestry conditional on phenotype data using an ascertained ancestral recombination graph (ARG).

Main Methods:

• An ascertained ancestral recombination graph (ARG) was developed to model chromosome sharing.
• A gene mapping algorithm utilizing lod scores and permutation testing was defined.
• A Hidden Markov algorithm was employed to handle incomplete penetrance and varying marker frequencies.

Main Results:

• The proposed method allows for exact computation of lod scores and p-values without Monte Carlo approximations.
• The algorithm effectively handles unphased genotype data, incomplete penetrance, and varying allele frequencies.
• Performance was validated through simulations and a real breast cancer case-control study.

Conclusions:

• The ascertained ARG provides a powerful framework for gene mapping in phenotyped samples.
• This novel algorithm offers an efficient and accurate approach for genetic linkage analysis.
• The method has potential applications in complex disease gene discovery.