Inference of gene relations from microarray data by abduction

We describe an application of Abductive Logic Programming (ALP) to the analysis of an important class of DNA microarray experiments. These experiments measure di erences in expression levels of whole genomes in di ering environmental conditions and/or after deletion or overexpression of one or more genes. Their aim is to obtain insights about gene interactions and gene pathways. We develop an ALP theory that provides a simple and general model of how gene interactions can cause changes in observable expression levels of genes. Input to the procedure are the observed microarray results; output are hypotheses about possible gene interactions that explain the observed e ects. A key feature of the model are parameters that encode di erent biological assumptions and provide a means of constraining the search for possible hypotheses. We have applied and evaluated our approach on microarray experiments on M.tuberculosis and on S.cerevisiae (yeast). Comparison of inferred hypotheses against known gene regulation networks and known gene functions in the biological literature provide a form of independent validation of the model