Protein Folding Recognition Based on Particle Swarm Optimization and Optimal Interval Distribution Machine

The study of protein folding can provide important support for the early warning of gene diseases and the design of protein target drugs. In this paper, combined particle swarm optimization algorithm (PSO) with multi-classification optimal interval distribution machine (mcODM), a novel protein folding recognition model called pso-mcODM is proposed, in which both the first and second order statistics of the minimized interval in the objective function are considered, and the non-convex and non-smooth objective functions are solved by the method of random mirror proximal descent. PSO performs global optimization of three parameters in mcODM loss function. Finally, a new adaptive parameter optimization classification model pso-mcODM is constructed.