A new branch and bound method for the protein folding problem under the 2D-HP model.

The protein folding problem is a fundamental problem in computational molecular biology and biochemical physics. The previously best known branch and bound method for the protein folding problem may find optimal or near-optimal energy structure from the benchmark sequences, but the total computation time is rather lengthy because it usually needs to run a great deal of simulating tests or else lack of accuracy. In this paper, we develop a new branch and bound method for the the protein folding problem under the two-dimensional HP model to overcome the mentioned drawbacks. By using benchmark sequences for evaluation, we demonstrate that the performance of our method is superior than previously known methods. Moreover, our method is a simple, flexible and easily implemented one for the protein folding problem.