Information- and graph-theoretic analysis of bacterial two component system

The bacterial two-component system is the major signal transduction system consisting of two genes, a histidine kinase (HK) and a response regulator (RR). The mechanism of how these simple systems maximize highly specific signal array and minimize irrelevant crosstalks has not been well characterized. In particular, the underlying biochemical mechanism of functional HK-RR pairs is yet to be understood since HK and RR genes share high level of sequence similarity. In this paper, we show, using an information- and graph-theoretic approach, that there are non-conserved, but highly co-evolving amino acid residue pairs in functional HK-RR pairs, but not in potentially non-functional HK-RR pairs. This finding can potentially help us to understand how HK and RR pairs and can lead us to develop a computational method to predict the functional HK-RR pairs and a guidance system for the mutation experiment design for a TCS.