Representing inter-residue dependencies in protein sequences with probabilistic networks

A new method for representing a local region of a protein sequence as a probabilistic network is proposed. The method produces, from a large number of examples of a local region, a network which describes dependencies that exist among amino acid residues in the region. The network is constructed using the greedy-search algorithm based on the minimum description length (MDL) principle. In our experiments, we construct a probabilistic network of the EF-hand motif domain in calcium-binding proteins. Experimental results show that our method provides a visual aid to understanding the inter-residue dependencies of those regions using a probabilistic network, and the network captures several important features which are peculiar to the motif.