ICGC Data Portal

ICGC Data Portal provides unified access to cancer genomics datasets for characterization and integrative analysis of somatic mutations, copy number alterations, structural rearrangements, gene expression profiles, microRNAs, DNA methylation patterns, exon junctions, and controlled germline variation across multiple cancer projects.

Key Features:

• BioMart integration: Powered by the BioMart software to integrate distributed member-institution databases into a unified queryable resource.
• Distributed data management: Allows member institutions to manage local databases while integrating those resources into the portal.
• Data scope and provenance: Hosts data from 24 cancer projects, including contributions from The Cancer Genome Atlas (TCGA), Johns Hopkins University, and the Tumor Sequencing Project, encompassing 3478 genomes across 13 cancer types and subtypes and targeting characterization of genomic abnormalities in 50 cancers.
• Supported data types: Provides open-access somatic mutations, copy number alterations, structural rearrangements, gene expression profiles, microRNAs, DNA methylation patterns, and exon junctions, with controlled access to simple germline variations.
• Programmatic access and querying: Offers application programming interfaces (APIs) for programmatic access and supports construction of complex queries across multiple datasets.

Scientific Applications:

• Genomic aberration characterization: Characterization of somatic and structural genomic abnormalities across diverse cancer types.
• Integrative multi-omic analysis: Integrative analyses combining somatic mutations, copy number alterations, structural rearrangements, gene expression, microRNAs, DNA methylation, and exon junction data.
• Pan-cancer and cross-project comparisons: Pan-cancer comparative analyses and cross-referencing of datasets from TCGA and other contributing projects.
• Germline–somatic interaction studies: Controlled-access simple germline variation analyses to investigate germline contributions and interactions with somatic events.
• Cohort-level computational studies: Automated cohort-level data retrieval and analysis via APIs for large-scale computational oncology studies.

Methodology:

The portal is powered by BioMart; member institutions maintain local databases that are integrated into a unified access layer, and APIs support programmatic access and construction of complex cross-dataset queries.