MDR (Multifactor Dimensionality Reduction) is a data mining strategy for detecting and characterizing nonlinear interactions among discrete attributes (e.g. SNPs, smoking, gender, etc.) that are predictive of a discrete outcome (e.g. case-control status).

MDR is a genetic and nonparametric model-free data mining alternative to logistic regression for characterizing and detecting nonlinear interactions among discrete genetic and environmental attributes.

The MDR method combines attribute selection, attribute construction, and classification with cross-validation and permutation testing to provide a comprehensive and powerful approach to detecting nonlinear interactions.