Abstract: In this paper，a multiple sequence alignment is presented based on improved ant colony algorithm.Some changes have done in the algorithm including the renewal mode of the pheromone，the select means of the characters，the search strategy of ants on the trip between ant nest and food，the random distributing of the ants beginning sequence.The results show that stagnation behavior of basic ant colony mutiple sequence alignment algorithm is avoided and a good ability of searching better solution in the last runs of the improved algorithm is achieved.

Key words: ant colony algorithm, multiple sequence alignment, pheromone

摘要： 提出一种基于改进蚁群算法的多序列比对方法。该算法改变了信息素的更新方式、字符的选择方法、蚂蚁在蚁巢和食物之间往返搜索以及随机分配蚂蚁开始序列等。实验结果表明，改进后的算法不仅有效地克服了基本蚁群多序列比对算法中的停滞现象，而且即使在运行的后期，仍然能以极大的概率搜索较好解。

关键词: 蚁群算法, 多序列比对, 信息素