A new efficient method to detect genetic interactions for lung cancer GWAS

Area of Science:

• Genetics
• Computational Biology
• Biostatistics

Background:

• Genome-wide association studies (GWAS) face computational challenges in identifying single nucleotide polymorphism (SNP) interactions for survival analysis.
• Existing methods are limited in detecting complex genetic interactions influencing disease onset age.

Purpose of the Study:

• To develop a novel algorithm, Efficient Survival Multifactor Dimensionality Reduction (ES-MDR), to address computational burdens in detecting SNP interactions for survival analysis.
• To identify significant genetic interactions associated with age of disease onset.

Main Methods:

• Developed ES-MDR using Martingale Residuals for survival outcome estimation.
• Implemented Quantitative Multifactor Dimensionality Reduction to identify significant interactions.
• Evaluated ES-MDR on simulation data for type I error rate and power, and applied it to OncoArray-TRICL lung cancer data.

Main Results:

• ES-MDR demonstrated reduced computational workload and allowed covariate adjustment in simulations.
• Analysis of OncoArray-TRICL data identified significant one-way and two-way models.
• A single-base deletion in the noncoding region of BRCA1 was the top marker predicting lung cancer age of onset (HR 1.24, P = 3.15 × 10⁻¹⁵).

Conclusions:

• ES-MDR is an efficient algorithm for identifying genetic interactions in survival outcome prediction.
• The method successfully identified key genetic markers associated with age of lung cancer onset.
• This approach enhances the ability to model complex genetic influences on survival.