ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking

TLDR

Unsupervised class discovery is a valuable technique in cancer research, and consensus clustering provides quantitative and visual stability evidence for estimating the number of intrinsic classes. ConsensusClusterPlus implements the consensus clustering method in R and adds new functionality and visualizations. It extends the method with item tracking, item-consensus, and cluster-consensus plots. The added features give users detailed information that enables more specific decisions in unsupervised class discovery. ConsensusClusterPlus is open source R software under GPL‑2, available via Bioconductor (http://www.bioconductor.org/), with contact [e-mail] and supplementary data online.

Abstract

Abstract Summary: Unsupervised class discovery is a highly useful technique in cancer research, where intrinsic groups sharing biological characteristics may exist but are unknown. The consensus clustering (CC) method provides quantitative and visual stability evidence for estimating the number of unsupervised classes in a dataset. ConsensusClusterPlus implements the CC method in R and extends it with new functionality and visualizations including item tracking, item-consensus and cluster-consensus plots. These new features provide users with detailed information that enable more specific decisions in unsupervised class discovery. Availability: ConsensusClusterPlus is open source software, written in R, under GPL-2, and available through the Bioconductor project (http://www.bioconductor.org/). Contact: [e-mail] Supplementary Information: Supplementary data are available at Bioinformatics online.