A Knowledge-Based Artificial Bee Colony Algorithm for the 3-D Protein Structure Prediction Problem

The prediction of protein structures is one of the most challenging problems in Structural Bioinformatics. The challenge relies on the combinatorial explosion of plausible shapes, where the conformational space grows exponentially as the number of amino acids in protein sequences increases. Many computational strategies were proposed over the last decades. Nevertheless, the problem remains an open challenge. In this paper, we present a first principle method with database information for the 3-D protein structure prediction problem. We do so by exploring swarm intelligence concepts by designing modified versions of the artificial bee colony algorithm to address the concerned problem and assess the real potential of the method against an extremely complex problem. The methods also take advantage of structural knowledge from the Protein Data Bank to better guide the search and restrict the conformational space. To validate our computational strategies, we tested them on a set of eight protein sequences. Predicted structures were analyzed regarding root-mean-square deviation and global distance total score test. Obtained results for the final algorithm outperformed its previous version, demonstrating the importance of adapting the algorithm to deal with the particularities of the problem. Further, the achieved results are topologically compatible with the experimental correspondent, thus corroborating the promising performance of the method.