Monte Carlo methods for linkage analysis of two-locus disease models

Area of Science:

• Genetics
• Statistical genetics
• Computational biology

Background:

• Parametric linkage analysis for qualitative traits typically uses single-locus models.
• Two-locus models offer greater power for detecting linkage but are computationally intensive.
• Current methods limit two-locus analysis to single markers, hindering comprehensive gene mapping.

Purpose of the Study:

• To introduce a computationally efficient Markov chain Monte Carlo (MCMC) method for two-locus parametric linkage analysis.
• To enable mapping of traits to multiple markers, increasing analytical power.
• To validate the MCMC method using an alcohol dependence dataset and simulations.

Main Methods:

• Developed a Markov chain Monte Carlo (MCMC) algorithm for two-locus lod-score analysis.
• The MCMC algorithm's computational complexity is linear with the number of markers.
• Applied the method to an alcohol dependence dataset and conducted simulation studies.

Main Results:

• The MCMC method provides lod-score estimates comparable to exact analysis but with significantly reduced computation time.
• Analysis of an alcohol dependence dataset (105 pedigrees) demonstrated the method's efficiency and accuracy.
• Simulation studies confirmed that incorporating additional markers substantially increases linkage detection power.

Conclusions:

• The proposed MCMC method significantly enhances the feasibility and efficiency of two-locus linkage analysis.
• This approach facilitates more powerful and comprehensive genetic mapping of complex traits.
• The findings support the routine use of two-locus analysis for disease gene discovery.