基于随机聚类方法建模的序列分析

摘要

大数据下的系统发育估计是一个组合优化问题,在有限计算时间内,现有算法很难为大量序列数据的分析提供最优解.基于前人启发式算法,提出了一种系统发育树随机聚类建树方法,可在较短时间内为系统发育过程产生的大规模序列数据提供所有具有进化意义的解及最优解,以揭示发育过程中的序列进化关系.实验结果表明,该随机聚类方法是行之有效的,对生物计算及系统发育相关领域研究具有积极意义.%Large phylogeny estimation is a combinatorial optimization problem that no future computer will ever be a-ble to solve exactly in practical computing time. Here,a tree constructing approach has been reported,the random clustering method,involving several pruning of trees that are used to provide the optimal solution and near-optimal solution with evolutionary significances,to reveal the complete evolutionary relationships based on basis of previous studies. The experiments show the correctness and efficiency of our method,and the significances to biocomputing and phylogenetic analysis.