GEAMM v.1.4: a versatile program for mixed model analysis of gene expression data

Abstract

[EN] This note presents the program geamm v.1.4 (Gene Expression Analysis with Mixed Models), a versatile software program to analyse appropriately normalized gene expression data by a heteroskedastic mixed model, and allowing for both discrete and continuous effects in model. Although the analysis of gene expression data is an area of intensive software development, available programs are focused on the comparison between discrete¿type effects1 (e.g., health vs. cancerous cells2), whereas continuous effects like age3 cannot be easily accommodated. To our best knowledge, this is the first microarray¿specific program allowing for an easy and flexible modelling of continuous effects on gene expression analysis. Mixed model equations are implemented under Bayesian inference and solved through Gibbs sampling. The program, user's guide and example files can be downloaded free of charge from http://www.bdporc.irta.es/Publicacions/GEAMM.zip. A user's guide is also available as Appendix S1.