A Protein Structure Refinement Rotocol Using Multi-objective Particle Swarm Optimization

Protein structure refinement is an important procedure to narrow the gap between the final prediction structure and the native structure.The 3D structure of a protein is of great significance for understanding its biological functions and the interaction of biological molecules,and the structure of proteins is also the base for new drug design and medical development.In the field of protein structure refinement,there are three unsolved problems.The first is the imperfect energy function,which leads to the fact that the structure with the lowest energy is not always near the native structure.The second is the lack of effective searching algorithm due to the huge conformational space.And the third is difficulty in selecting the best model from the candidate set.To solve these problems,an artificial intelligence-based protein Refinement method called AIR is proposed,which is constructed using a multi-objective particle swarm optimization(PSO)protocol.The basic motivation of AIR is to solve the bias problem caused by minimizing only a single energy function due to the diversity of different protein structures.The fundamental idea of our method is to use multiple energy functions as multi-objectives in an effort to correct the potential inaccuracy from a single function.In our method,each protein structure is considered as a particle in the protocol.With the process of structure refinement,the particles move around.The quality of particles(structures)in each iteration is evaluated by three energy functions,and the non-dominated particles are put into a set called Pareto set.After the iteration converges,the particles from the Pareto set are screened and part of the top solutions are outputted as the final refined structures.The existing 1-D optimization of a single energy function may take the structure too far away without a constraint.On the other hand,the multi-objective energy function optimization strategy designed in the AIR protocol provides a different constraint view of the structure,by extending the 1-D optimization to a new 3-D space optimization driven by the multi-objective PSO engine.To verify the validity of AIR,a series of tests on the CASP11 and CASP12 refinement targets and the comparison with two other refinement methods are conducted,and the results turn to be very promising in terms of RMSD,TM-score and other evaluation criteria.The AIR is available online at: www.csbio.sjtu.edu.cn/bioinf/AIR/.