Supplement to

Molecular Decomposition of Complex Clinical Phenotypes using Biologically Structured Analysis of Microarray Data

Abstract

Today, the characterization of clinical phenotypes by distinctive gene expression patterns obtained from Microarrays is widely used in clinical research. However, it is still a challenging task, if the clinical phenotype is complex from the molecular point of view. The same clinical phenotype can be caused by various molecular disorders, such that one observes different characteristic expression patterns in different patients.

In this paper we describe a novel algorithm called Structured Analysis of Microarrays (StAM), which accounts for molecular heterogeneity of complex clinical phenotypes. In addition to the expression data our algorithm also exploits functional annotations from the Gene Ontology database. On three publicly available cancer-related microarray data sets, we show that StAM achieves state-of-the-art prediction accuracy, while at the same time uncovering potential molecular disease sub-entities and associating them to biological processes.

Keywords: microarrays, gene expression profiles, class prediction, Gene Ontology, molecular symptom, complex phenotype.

Berlin Center for Genome based Bioinformatics      Max-Planck-Institute for Molecular Genetics, Berlin Computational Diagnostics Group