CrossICC

CrossICC performs iterative consensus clustering of cross-platform gene expression data as an R package to identify robust cancer subtypes without requiring between-dataset normalization.

Key Features:

• Iterative Consensus Clustering: Derives optimal gene sets and cluster numbers from a consensus similarity matrix generated via consensus clustering techniques.
• Cross-Platform Compatibility: Operates on multiple cross-platform gene expression datasets without requiring between-dataset normalization or explicit batch effect adjustment.
• Visualization and Evaluation: Provides functions for visualizing identified cancer subtypes and evaluating subtyping performance across datasets.
• Cancer-Specific Analysis Methods: Incorporates methods tailored for cancer-related analyses to support subtype characterization and downstream interpretation.

Scientific Applications:

• Cancer Subtype Discovery: Identification of robust and reproducible cancer molecular subtypes from heterogeneous gene expression datasets.
• Translational and Personalized Medicine: Generation of consistent subtypes to support translational research, targeted therapy development, and patient-specific treatment strategy studies.

Methodology:

Iterative refinement of gene signatures and cluster numbers using consensus clustering to produce a consensus similarity matrix from cross-platform gene expression data.