A cooperative fast annealing coevolutionary algorithm for protein motif extraction  

作　　者：CHEN Chao TIAN YuanXin ZOU XiaoYong CAI PeiXiang MO JinYuan 

机构地区：[1]School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China

出　　处：《Chinese Science Bulletin》2007年第3期318-323,共6页

基　　金：the National Natural Science Foundation of China (Grant Nos. 20475068 and 20575082);the Natural Science Foundation of Guangdong Province (Grant No. 031577);the Scientific Technology Project of Guangdong Province (Grant No. 2005B30101003)

摘　　要：By integrating the cooperative approach with the fast annealing coevolutionary algorithm (FAEA), a so-called cooperative fast annealing coevolutionary algorithm (CFACA) is presented in this paper for the purpose of solving high-dimensional problems. After the partition of the search space in CFACA, each smaller one is then searched by a separate FAEA. The fitness function is evaluated by combining sub-solutions found by each of the FAEAs. It demonstrates that the CFACA outperforms the FAEA in the domain of function optimization, especially in terms of convergence rate. The current algorithm is also applied to a real optimization problem of protein motif extraction. And a satisfactory result has been obtained with the accuracy of prediction achieving 67.0%, which is in agreement with the result in the PROSITE database.By integrating the cooperative approach with the fast annealing coevolutionary algorithm （FAEA）, a so-called cooperative fast annealing coevolutionary algorithm （CFACA） is presented in this paper for the purpose of solving high-dimensional problems. After the partition of the search space in CFACA, each smaller one is then searched by a separate FAEA. The fitness function is evaluated by combining sub-solutions found by each of the FAEAs. It demonstrates that the CFACA outperforms the FAEA in the domain of function optimization, especially in terms of convergence rate. The current algorithm is also applied to a real optimization problem of protein motif extraction. And a satisfactory result has been obtained with the accuracy of prediction achieving 67.0%, which is in agreement with the result in the PROSITE database.

关 键 词：全局优化 合作法 快速退火 进化算法 蛋白质 超二级结构 功能域 

分 类 号：Q518.1[生物学—生物化学]