Two hypotheses and test assumptions based on Quantum-behaved Particle Swarm Optimization (QPSO)

In this paper, two hypotheses are proposed to explain protein folding, aiming at this hypotheses, an improved 3D Off-lattice model and Quantum-behaved Particle Swarm Optimization based protein folding algorithm (c-QPSO) for predicting the protein folding structure are also proposed. The results are sufficiently lowest energy better than the results determined by the other algorithms for verifying the two hypotheses. For Fibonacci sequence with size 13, 21 and 34, our result is roughly twice times than that of ELP and 1.5 times for sequence with size 55. It can also be seen that c-QPSO algorithm takes significantly less time than PSO method. For real protein sequences, the structures predicted by c-QPSO can approximately simulate the real protein to some extent. Experiments show that the two hypotheses is partially correct, and the improved 3D Off-lattice model is beneficial for c-QPSO algorithm to reduce the computation time and easily get the 3D coordinate of each amino acid. Some proteins fold faster than they elongate, and it is reasonable to assume that nascent chains can adopt secondary or tertiary structures cotranslationally.