A rule-based framework for gene regulation pathways discovery

We present a novel approach for rule discovery in the area of gene regulation. The method is based on supervised machine learning and is designed to reveal relationships between transcription factors and gene expression. As a representation of the gene regulatory circuit we have chosen a special form of the IF-THEN rules associating predefined features, taken here as a generalized idea of transcription factor binding sites in gene regulatory regions, with the corresponding gene expression profiles.