Application Search Of Annealing Evolution Algorithm In Biology Sequence Alignment

Comparison is the most common method in the scientific research, that through the comparision of research object with each other relatively look for the characteristic that target might possess. Sequence alignment is the most frequently and classical research means of bioinformatics. Its meaning rest with analyzing sequence comparability from the level of nucleic acid and amino acid, conferring its relation of structural function and evolution, the foundation of gene discern, molecule evolution and origin of life study.Have introduced the current situation of study on algorithm that pairwise sequence and multiple sequence alignment in bioinformatics in this text. Than based on simulated annealing algorithm and genetic algorithm, propose to an annealing evolution algorithm in biology sequence alignment. The algorithm complexity of multiple sequence increases in the exponent law, is a NP problem, according to the characteristic we adopt the algorithm belongs to iterative algorithm that combine with simulated annealing algorithm and genetic algorithm, treatment NP problem that this kind of method can be very good. Because genetic algorithm can be easy to cause early-maturing to disappear question, so as to evolution be unable to get optimum to solve. A lot of multiple sequence alignment based on genetic algorithm have this shortcoming more than the algorithm right. According to this characteristic, through introducing simulated annealing algorithm, utilize simulated annealing algorithm, accepting criterion (Metropolis criterion ) keep colony variety of individual to solve early-maturing to disappear question. Simulated annealing has two-way search ability that probability jumps suddenly, it is apt to jump out the trap of some extreme value, the overall optimization that can also guarantee to be searched for. Introducing simulated annealing algorithm in the genetic algorithm tactics, not only enrich the genetic algorithm search behavior, avoid disappear early-maturing, but also can utilize strong the overall optimization of genetic algorithm running side by side of to search for ability.This text carried on a series of data test finally, passing the test result, we can see this algorithm is improved in biological sensitiveness and operation efficiency to some extent compared with the traditional algorithm.