Simulation of electric charge distribution of all proteins in Saccharomyces cerevisiae proteome

It is widely believed that the protein structures are determined by the combination of various physical interactions, among which the electrostatic interaction between electric charges is one of the most well-defined physical factors. Therefore, the electrostatic effects have been studied for various individual proteins so far [1, 2]. However, the analysis of the electric charge distribution of total proteins in a proteome has little been studied yet. The electric charge distribution of total sequences in a proteome is the result of evolutionary processes, and there must be two kinds of evolutionary pressures on the variation of charges in sequences: On one hand, the sequences of electric charges are randomized by mutations of amino acid sequences. On the other hand, some of charged residues are preserved because of their functional and structural importance. Thus, interesting questions arise how random is the sequences of charged residues in a proteome and whether the attractive interaction is dominant for the formation of globular proteins. In this work, we calculated the autocorrelation function of electric charges in all amino acid sequences of the Saccharomyces cerevisiae proteome. Furthermore, we carried out simulations of amino acid sequences so that the autocorrelation function of electric charges can be reproduced by simple rule of the generation of charged residues. The results indicated that the repulsive interaction was surprisingly dominant in the total proteomes and this characteristic feature of a proteome is partially reproducible, assuming the local positive correlation of charged residues, but the long range repulsion could not be reproduced easily.