Inserting Co-Evolution Information from Contact Maps into a Multiobjective Genetic Algorithm for Protein Structure Prediction

Protein structure prediction (PSP) can be described as a multiobjective optimization (MO) problem since the energy function involves potentially conflicting terms to be simultaneously optimized. During the last three CASP editions (10th, 11th, and 12th), promising results were achieved with the introduction of co-evolution information, in the form of residues contact maps, in methodologies for PSP. In this paper, a residue-residue contact map potential is introduced into the evaluation function of the GAPF program, and it is optimized using a MO strategy. The Aggregation Tree (AT) method is applied to group in separated objectives the energetic potentials that compose the GAPF's evaluation function. The results are compared with those obtained from two consolidated PSP methods, QUARK and MEAMT.