Bayesian microarray one-way anova and grouping cell lines by equal expression levels

Motivation: In the exploding field of gene expression techniques such as DNA microarrays, there are still few general probabilistic methods for analysis of variance. Linear models and anova are heavily used tools in many other disciplines of scientific research. The usual F -statistic is unsatisfactory for microarray data, which explores many thousand genes in parallel, with few replicates. Results: We present a probabilistic parametric framework, which produces three potential oneway anova statistics that separate genes which are differently regulated across several treatment conditions from those that have equal regulation. The statistics have different features and are evaluated using both real and simulated data. Our statistic B2 generally shows the best performance, and is extended for use in an algorithm that groups significantly regulated genes by equal expression levels. Availability: The methods presented are implemented in the freely available statistical language R. They are available at http://www.math.uu.se/staff/pages/?uname=ingrid. Contact: ingrid@math.uu.se Supplementary Information: (Web reference?)