期刊名称：Inteligencia Artificial : Ibero-American Journal of Artificial Intelligence
印刷版ISSN：1137-3601
电子版ISSN：1988-3064
出版年度：2012
卷号：15
期号：49
页码：1-12
出版社：Spanish Association for Intelligence Artificial
摘要：Search methods can find state sequences that relate two or more states using a set of allowed transitions. Evolutive algorithms perform search using an aptitude function and use stochastic operators for exploring multiple solutions at once. In bioinformatics, pathway search between to compounds is a common task. It is particularlly important in searching metabolic relationships in a set of compounds grouped with data mining techniques. This work proposes an evolutive algorithm to find metabolic pathways between two compounds selected from different groups found with a self-organizing map-like model. We describe the operators and the fitness function used, we study the effect of the mutation rate of the proposed algorithm and compare its performance with two traditional search methods.
