The MDMR analysis method allows predictor variables collected on the samples to be related to variation in pair-wise distance values reflected in a distance matrix. In addition to eliminating the need for reducing the dimensions of the matrix, this multivariate method can be used to assess relationships between data for constructing the matrix and additional information collected on the samples under study, and to analyze individual data points or groups of data points identified in different ways. MDMR provides a formal statistical test, rooted in traditional linear models, of how independent variables are associated to the variation present in a pair-wise distance matrix. This program, written by Ondrej Libiger and Matt Zapala is based on DISTLM by Marti J. Anderson.
For more information: