An Ant Colony Optimization Algorithm for the 2D HP Protein Folding Problem

Abstract

The prediction of a protein’s conformation from its amino-acid sequence is one of the most prominent problems in computational biology. Here, we focus on a widely studied abstraction of this problem, the two dimensional hydrophobic-polar (2D HP) protein folding problem. We introduce an ant colony optimisation algorithm for this NP-hard combinatorial problem and demonstrate its ability to solve standard benchmark instances. Furthermore, we empirically study the impact of various algorithmic features and parameter settings on the performance of our algorithm. To our best knowledge, this is the first application of ACO to this highly relevant problem from bioinformatics; yet, the performance of our ACO algorithm closely approaches that of specialised, state-of-the methods for 2D HP protein folding.KeywordsLocal SearchSequence PositionProtein Structure PredictionBenchmark InstanceFolding ProcessThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.