[Detecting gene-gene/environment interactions by model-based multifactor dimensionality reduction].

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• Detecting gene-gene and gene-environment interactions is crucial in genetic studies.
• Classical multifactor dimensionality reduction (MDR) is a common method but has limitations.
• Model-based MDR (MB-MDR) offers potential improvements.

Purpose of the Study:

• Introduce and summarize the model-based multifactor dimensionality reduction (MB-MDR) method.
• Illustrate the application and procedure of MB-MDR using an example.
• Compare MB-MDR with classical MDR in terms of statistical power and applicability.

Main Methods:

• MB-MDR merges multi-locus genotypes into a one-dimensional construct.
• The method is implemented and demonstrated using the R program.
• It adjusts for marginal effects of factors and confounders.

Main Results:

• MB-MDR exhibits higher statistical power than classical MDR, especially in noisy datasets.
• It demonstrates greater flexibility in handling both binary and quantitative traits.
• The method is suitable for genetic studies with small sample sizes.

Conclusions:

• MB-MDR is a powerful and versatile method for analyzing gene-gene/environment interactions.
• It offers advantages over classical MDR in statistical power and trait adaptability.
• MB-MDR is a valuable tool for genetic association studies.