Abstract

We demonstrate the Binary Identified Contact Domain (BICD) method for predicting binding sites for proteins that starts with amino-acid sequences of several species, finds regions that are evolutionarily conserved, and weights regions by their local hydrophilicity. The measure of conservation takes transversions as significant but transitions as insignificant, in keeping with a theory of original singlet codes. The BICD method correctly predicts several known binding sites for the BRCA2 protein with PABL2 and RAD51 as well as the binding sites of the MDM2-P53 complex.

Keywords: Amino acids, binary code, binding sites, codons, conservation, evolution, genetic code, triplet.