Research on optimal Multiple Sequence Alignment

Multiple Sequence Alignment(MSA) is one of the most fundamental problems in biology information. At present, with the rapid increase in the sequence quantity, there is an urgent need to run the optimized algorithm of MSA, the researchers also made a lot of the problem solution, such as dynamic programming, including the Needleman-Wunsch algorithm and Carrillo-Lipman algorithm, Feng-Doolittle algorithm and algorithm ClustalW of heuristic algorithm step method; the Barton & Sternberg algorithm of Iterative optimization method; HMMs of statistics and probability method and multiple sequence alignment algorithms in genetic algorithm of random optimization method, such as SAGA. In this paper, studied MSA optimization algorithm of genetic algorithm-based, which put forward an improved adaptive genetic algorithm, calibrated fitness value, and realized the community multiplicity in the evolution process, while the use of the most commonly used sigmoid function in structural neuronal activation function, carried on the optimization to overlapping rate and the rate of mutation, has realized misalignment auto-adapted adjustment of overlapping rate and mutation rate, sharpened the algorithm optimization ability and the stability.