PCaDB

PCaDB provides a harmonized repository and analytical resource for prostate cancer transcriptomics to support molecular characterization and prognostic signature evaluation.

Key Features:

• Extensive Data Collection: PCaDB integrates 77 transcriptomics datasets encompassing 9,068 patient samples for prostate cancer research.
• Data Harmonization and Standardization: The resource performs integration and standardization of transcriptomics data across cohorts to ensure consistency and comparability.
• Single-Cell RNA Sequencing (scRNAseq): The database includes an scRNAseq dataset derived from normal human prostates for cellular-resolution gene expression analysis.
• Prognostic Signatures: PCaDB incorporates 30 published prognostic signatures for evaluation and comparison.
• Analytical Methods and Machine Learning Support: The resource provides analytical methods to support data mining and development or assessment of machine learning models for diagnostic and prognostic studies.

Scientific Applications:

• Molecular characterization of heterogeneity: Use harmonized transcriptomics datasets to investigate molecular diversity within prostate cancer.
• Development and validation of diagnostic and prognostic signatures: Evaluate published signatures and derive or compare new signatures using aggregated cohorts.
• Machine learning model development and evaluation: Train and assess machine learning models for prognosis and diagnosis using the compiled transcriptomics and signature data.

Methodology:

Integration and standardization of public prostate cancer transcriptomics datasets across cohorts, inclusion of a single-cell RNA sequencing dataset from normal human prostates, and incorporation of 30 published prognostic signatures.