Artikel
Genome-wide Interaction Analysis Guided by A Priori Information
Suche in Medline nach
Autoren
Veröffentlicht: | 2. September 2009 |
---|
Gliederung
Text
Complex diseases are caused by interacting genetic and environmental factors. Due to computational burden, genome-wide association studies (GWAS) are typically limited to single-marker analysis. We present an approach for genome-wide interaction analysis that overcomes the computational issue by prioritizing SNPs for interaction analysis using a priori information. Sources of information can be biological relevance (gene location, function class ...) or statistical evidence (single marker association at a moderate level). We present a respective software product that implements different approaches to joint analysis of multiple SNPs (full modelling of marginal and interaction effects, as well as explicit testing for interaction with a log-linear model). The software implements various methods to account for multiple testing and to judge genome-wide significance. In particular, genome-wide Monte-Carlo simulations are feasible. We present the results from an application to a GWAS.