Preprocessed Bulk RNA-seq Datasets for CAMDA Health Privacy Challenge 2025

This record contains the preprocessed bulk gene-expression (RNA-seq) datasets (TCGA-BRCA and TCGA COMBINED) used in CAMDA Health Privacy Challenge 2025, and its accompanying paper, Community benchmarking of generative models for privacy-preserving synthetic bulk RNA-seq data generation. [details to be added soon] TCGA-BRCA.zip: Two files; preprocessed bulk RNA-seq breast cancer data and subtype information, respectively. TCGA-COMBINED.zip: Three files; preprocessed bulk RNA-seq of 10 cancer tissues and cancer type information, respectively and a held-out (preprocessed) bulk RNA-seq reference set to be used as auxiliary data. splits_indices.zip: Two files; YAML files required to split the datasets into 5 fold-cv subsets. License Creative Commons Attribution 4.0 International. This license applies to the preprocessed/derived dataset provided in this record. It does not grant additional rights to the underlying TCGA/GDC data and does not supersede applicable TCGA, NCI, NIH, or GDC policies. Data use notice Please read Data_Download_Notice.md before downloading or using these data. The notice describes the data provenance, applicable TCGA/GDC requirements, citation requirements, and restrictions on re-identification.