Estimation of 3D Protein Structure by Means of Parallel Particle Swarm Optimization

This paper presents the Algorithm of Particle Swarm Optimization (PSO) implemented in a Distributed Computing Environment. The main objective of this PSO is to calculate the Protein 3D Structure reducing the computational time to carry out this task; this problem is also known as Protein Folding Problem (PFP). The parallel PSO works on a real conformation, considering structural restriction of the protein, where the conformation uses a representation of torsion angles of the skeleton and the side chains, applying the sequence of amino-acid of the protein for the prediction of 3D structure of minimum energy. In order to calculate the energy of the protein conformation, the energy empirical function ECEPP/3 is used. This program was implemented for running in a cluster with the libraries MPI for the processor communication. The quality of the results on the testing peptide (leu-enkephalin) is compared with other techniques reported in literature, and also the PSO is used to predict the structure of unknown proteins.