一种快速的网络Motif检测工具
An Faster Network Motif Detection Tool
摘要:
网络模体提供了一种揭示大多数复杂网络的基本构建块的方法。这个任务通常需要很高的计算机处理,特别是对于具有5个或更多顶点的模体。本文提出了一种扩展的方法,具有以下特点:(i)搜索图案多达6个顶点,(ii)多线程处理,和(iii)新的枚举算法具有较低的复杂度。计算模体的算法利用哈希表解决了$O(1)$中的同构问题。并发线程评估不同的图。枚举算法具有较小的计算复杂性。该实验显示出比文献中的其他方法更好的性能,使生物信息学研究人员能够有效地识别大小为3,4,5和6的基序。
英文摘要:
Network motif provides a way to uncover the basic building blocks of most complex networks. This task usually demands high computer processing, specially for motif with 5 or more vertices. This paper presents an extended methodology with the following features: (i) search for motifs up to 6 vertices, (ii) multithread processing, and a (iii) new enumeration algorithm with lower complexity. The algorithm to compute motifs solve isomorphism in $O(1)$ with the use of hash table. Concurrent threads evaluates distinct graphs. The enumeration algorithm has smaller computational complexity. The experiments shows better performance with respect to other methods available in literature, allowing bioinformatic researchers to efficiently identify motifs of size 3, 4, 5, and 6.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。