Abstract
The restricted partition method (RPM) is a partitioning algorithm for examining multi-locus genotypes as (potentially non-additive) predictors of a quantitative trait. The motivating application was to develop a robust method to examine quantitative phenotypes for epistasis (gene-gene interactions), but the method can be applied without modification to gene-environment interactions. Simulation results indicate that the method provides an efficient way to identify loci contributing epistatically to a quantitative trait, even if the loci have no single locus effects. Statistical significance can be estimated through permutation testing. An example using real data involving the metabolism of a chemotherapy drug is included for illustration. Although the examples in this article involve 2-locus interactions, the RPM is computationally feasible for the analysis of more than two loci or factors.
| Original language | English |
|---|---|
| Pages (from-to) | 141-152 |
| Number of pages | 12 |
| Journal | Genetic Epidemiology |
| Volume | 27 |
| Issue number | 2 |
| DOIs | |
| State | Published - Sep 2004 |
Keywords
- Methodology
- Pharmacogenetics
- Statistics