Global optimization of a three-dimensional HP protein

The conformational space annealing (CSA) method is an efficient, global optimization algorithm for sampling low-lying local-minimum-energy conformations of a physical system. In this work, the CSA method is applied to the three-dimensional HP lattice model of a protein, where a conformation is defined as a self-avoiding chain on the three-dimensional cubic lattice. The 27-residue chain with a particular sequence H2PHP2HPHPHPH2P3HP2H3P2HP is considered, for which compact conformations filling a 3 x 3 x 3 cube have been exhaustively enumerated in an earlier work, and only one compact conformation with the lowest energy has been shown to exist. The CSA algorithm finds non-compact conformations with energies lower than any compact conformation, for several independent runs, proving that the global minimum energy conformation is not a compact structure.