Supervised inference of gene-regulatory networks

The method is based on a kernel approach accompanied with genetic programming. As a data source, the method utilizes gene expression time series for prediction of interactions among regulatory proteins and their target genes. The performance of the method was verified using Saccharomyces cerevisiae cell cycle and DNA/RNA/protein biosynthesis gene expression data. The results were compared with independent data sources. Finally, a prediction of novel interactions within yeast gene expression circuits has been performed.Results show that our algorithm gives, in most cases, results identical with the independent experiments, when compared with the YEASTRACT database. In several cases our algorithm gives predictions of novel interactions which have not been reported.In recent years, the inference of protein interaction networks from various sources of data has become an important topic of both systems and computational biology. Protein networks can be represented as a graph with vertices formed by proteins and edges connecting two proteins representing the relationship between them. The interaction can be either direct, where two or more proteins form a functional complex, or indirect – biochemical or regulatory. The biochemical interaction can be, for example, the participation of two enzymes catalyzing two successive biochemical reactions in a pathway. Regulatory interaction represents binding of a transcription factor to a promoter site, which initiates transcription of a particular gene precursor of a protein.The diversity of protein "interactions" implies also diverse types of data ranging from literature references and sequence database annotations, through biophysical and biochemical data to the data from microarray and proteomics experiments. The type of data predetermines also the type of interaction studied. In this paper we focus on the gene expression networks, where a regulator protein controls expression of a gene precursor of the corresponding protein.In