蛋白质网络聚类算法分析平台的设计与实现

蛋白质网络聚类是识别功能模块的重要手段,不仅有利于理解生物系统的组织结构,对预测蛋白质功能也具有重要的意义。针对目前蛋白质网络聚类算法缺乏有效分析软件的事实,本文设计并实现了一个新的蛋白质网络聚类算法分析平台ClusterE。该平台实现了查全率、查准率、敏感性、特异性、功能富集分析等聚类评估方法,并且集成了FAG-EC、Dpclus、Monet、IPC-MCE、IPCA等聚类算法,不仅可以对蛋白质网络聚类分析结果进行可视化,并且可以在不同聚类分析指标下对多个聚类算法进行可视化比较与分析。该平台具有良好的扩展性,其中聚类算法以及聚类评估方法都是以插件形式集成到系统中。

Design and implementation of an analysis platform for clustering in protein interaction networks

Clustering in protein interaction network is an important means for the identification of functional modules, which not only helps to understand the organization of biological systems, but also is of great significance to predict protein function. According to the fact that the clustering algorithms of protein interaction network are lack of effective analysis software, we design and implement a new analysis platform, which is named ClusterE. Clus- terE is a scalable platform, in which a series of evaluation methods, such as recall, precision, sensitivity, specificity, p- value, and function enrichment, are implemented. Moreover, five clustering algorithms FAG -EC, Dpclus, Monet, IPC- MCE, IPCA are developed in this platform. Even more important, not only the clustering results are visualized, but also the comparison results of multiple algorithms can be shown in this platform. This platform has a good scalability for the clustering algorithms and evaluation methods are implemented as plug -ins.