Mining Plausible Patterns from Gene Expression Data

The discovery of biologically interpretable knowledge from gene expression data is one of the largest contempo- rary genomic challenges. As large volumes of expression data are being generated, there is a great need for auto- mated tools that provide the means to analyze them. How- ever, the same tools can provide an overwhelming number of candidate hypotheses which can hardly be manually ex- ploited by an expert. An additional knowledge helping to fo- cus automatically on the most plausible candidates only can up-value the experiment significantly. Background knowl- edge available in literature databases, biological ontolo- gies and other sources can be used for this purpose. In this paper we propose and verify a methodology that enables to effectively mine and represent meaningful over-expression patterns. Each pattern represents a bi-set of a gene group over-expressed in a set of biological situations. Original- ity of the framework consists in its constraint-based nature and an effective cross-fertilization of constraints based on expression data and background knowledge. The result is a limited set of candidate patterns that are most likely in- terpretable by biologists. Supplemental automatic interpre- tations serve to ease this process. Various constraints can generate plausible pattern sets of different characteristics.