DICE: Decoding Immune Cell Differentiation for Tumor-Type Prediction and Mechanism Interpretation
Abstract
large-scale multi-cohort cancer transcriptomic datasets to train and validate the DICE classification framework. Training data were obtained from The Cancer Genome Atlas (TCGA), initially comprising 33 cancer types. To guarantee data quality and statistical power, rigorous filtering criteria were implemented. This filtering process yielded 8,624 primary tumor samples across 24 cancer types.
Fields of Study & Subjects
Gene expression and cancer classification
Single-cell and spatial transcriptomics
Cancer Immunotherapy and Biomarkers
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 13 January 2026
- Document type
- Dataset