Graph-based consensus clustering for class discovery from gene expression data

摘要: Motivation: Consensus clustering, also known as cluster ensemble, is one of the important techniques for microarray data analysis, and particularly useful class discovery from data. Compared with traditional clustering algorithms, consensus approaches have ability to integrate multiple partitions different solutions improve robustness, stability, scalability parallelization algorithms. By can discover underlying classes samples in gene expression data. Results: In addition exploring a graph-based (GCC) algorithm estimate data, we design new validation index determine number To our knowledge, this first time which GCC applied Given pre specified maximum (denoted Kmax article), true according called Modified Rand Index. Experiments on indicate that (i) outperform most existing (ii) identify correctly real cancer datasets, (iii) biological meaning. Availability: Matlab source code available upon request Zhiwen Yu. Contact:yuzhiwen@cs.cityu.edu.hk cshswong@cityu.edu.hk Supplementary information: Supplementary are at Bioinformatics online.