Prediction Of Protein Structure Based On Tabu Search Algorithm

In natural protein, the final image is decided by ammo acid exclusively. A given amino acid series correspond a space structure on which biological functions are dependent extensively. Prediction protein structure is meaningful to interpret the relationship of protein structure and functions, and furthermore to perform protein complexity, mutation and medicament designing based on structure.Presently, the main approaches to measure protein structure are multi-dimension magnetic resonance technique and X-Ray diffraction method, which consume time and are constrained by experiment conditions. To predict protein structure utilizing computer technical is hence very meaningful. Many examples indicate that structure prediction problem is NP which can't be solved efficiently by a general method. And we don't know general regulations of natural protein. To solve the problem many scholars present simplified models which become a tool to research basic properties of protein folding.The work of this paper is to research predicting protein structure based on tabu search algorithm in research specific introduction is following .In the first part ,a memory tabu search algorithm is applied to solve 2-D off-lattice model of protein structure prediction problems. The performance of our algorithm is comparable to PERM algorithm. The model is simple and reflects the simple property of protein structure, the hydrophobic monomers form clusters of typically 4-5 particles and they are always flanked by polar monomers along the chain. Numerical tests illustrate that the memory tabu search algorithm is feasible and effective to protein structure prediction problems.In the second part, this paper proposes a new memory tabu search algorithm which is applied to continuous function optimization. Numerical results illustrate that this algorithm is feasible and effective and is very suitable for continuous global optimization.