Ant colony system: a cooperative learning approach to the traveling salesman problem
IEEE Transactions on Evolutionary Computation
Refinement on O atom positions for protein backbone prediction
BEBI'09 Proceedings of the 2nd WSEAS international conference on Biomedical electronics and biomedical informatics
Hi-index | 0.00 |
The protein side chain packing problem (PSCPP) is an essential issue for predicting structure in proteomics. PSCPP has been proved to be NP-hard. In this paper, we propose a method for solving PSCPP by transforming it to the graph clique problem, and then applying the ant colony optimization (ACO) algorithm to solve it. We build the coordinate rotamer library based on the pair of dihedral angles of backbones to reduce the required time. To evaluate the goodness of a solution of the ACO algorithm, we use a simple score function with four factors: disulfide bonds, intermolecular hydrogen bonds, charge-charge interactions and van der Waals interactions. The experimental results show that our score function is biologically sensible. We compare our computational results with the results of SCWRL 3.0 and the residue-rotamer-reduction (R3) algorithm. The accuracy of our method outperforms both of them.