COMSUC

COMSUC identifies consensus molecular subtypes across more than 30 cancer types by integrating genomic, transcriptomic, proteomic and other omics datasets with multiple clustering algorithms and consensus clustering approaches.

Key Features:

• Multi-omics integration: Aggregates genomic, transcriptomic, proteomic and other omics datasets to capture complex molecular profiles of cancer subtypes.
• Coverage of cancer types: Supports consensus subtype analysis across more than 30 types of cancers.
• Multiple clustering methods: Implements eight different clustering algorithms to generate diverse subtype partitions.
• Consensus clustering approaches: Utilizes three consensus clustering methods to reconcile discrepancies among clustering outcomes.
• Data flexibility: Accepts public reference datasets and private data as input for analysis.
• Collaborative exchange: Enables sharing of consensus subtype results using project IDs.
• Publishable outputs: Produces results formatted for inclusion in academic publications.

Scientific Applications:

• Consensus molecular subtype identification: Derives stable molecular subtypes from multi-omics data and multiple clustering results.
• Harmonization of clustering outcomes: Reconciles conflicting subtype assignments across algorithms and omics platforms to improve classification robustness.
• Tumor heterogeneity and precision oncology: Supports analysis of tumor heterogeneity and informs subtype-driven personalized treatment strategies.

Methodology:

Aggregating data from various omics platforms, applying diverse clustering algorithms to the integrated dataset, and integrating multiple clustering results using consensus methods to identify stable molecular subtypes.