Identification of metagenes and their interactions through large-scale analysis of Arabidopsis gene expression data

Many plant genes have been identified through whole genome and deep transcriptome sequencing and other methods; yet our knowledge on the function of many of these genes remains limited. The integration and analysis of large gene-expression datasets gives researchers the ability to formalize hypotheses concerning the functionality and interaction between different groups of correlated genes.

We applied the non-negative matrix factorization (NMF) algorithm to the AtGenExpress dataset which consists of 783 microarray samples (29 separate experimental series) conducted on the model plant Arabidopsis thaliana. We identified 15 metagenes, which are groups of genes with correlated expression. Functional roles of these metagenes are established by observing the enriched gene ontology (GO) categories using gene set enrichment analyses (GSEA). Activity levels of these metagenes in various experimental conditions are also analyzed to associate metagenes with stimuli/conditions. A metagene correlation network, constructed based on the results of NMF analysis, revealed many new interactions between the metagenes. Comparison of these metagenes with an earlier large-scale clustering analysis indicates many statistically significant overlaps.

This study identifies a network of correlated metagenes composed of Arabidopsis genes acting in a highly correlated fashion across a broad spectrum of experimental stimuli, which may shed some light on the function of many of the un-annotated genes.