TCGA whole‑genome scale

Cancer genome studies are moving beyond the 1% to 2% of DNA covered by exome tests and into full-genome maps from more than 8,000 matched tumor-normal cases. (aacrjournals.org) At the American Association for Cancer Research annual meeting in San Diego, researchers from the Broad Institute and Inocras reported a harmonized reanalysis of The Cancer Genome Atlas whole-genome data across more than 30 cancer types. The project consolidated more than 250 million variant calls from over 8,000 tumor-normal whole genomes. (businesswire.com) A whole genome is the complete instruction book in a cancer cell, not just the protein-coding pages. In the TCGA era, most studies used whole-exome sequencing, which samples only about 1% to 2% of the genome and can miss large rearrangements, copy-number shifts, and noncoding changes. (aacrjournals.org) The new analysis used an automated pipeline called CancerVision to detect single-letter mutations, small insertions and deletions, copy-number alterations, structural variants, and inherited mutations from the same whole-genome files. The team said it benchmarked those calls against Broad pipelines and official TCGA exome data in ovarian, thyroid, and glioblastoma cohorts. (aacrjournals.org) That broader view is already changing what researchers can measure. In one AACR abstract, the same dataset was used to map homologous recombination deficiency, a DNA repair failure tied to response to platinum drugs and PARP inhibitors, across more than 8,000 primary tumors. (aacrjournals.org) The authors reported homologous recombination deficiency patterns not only in breast and ovarian cancers, where current tests are common, but also in prostate, pancreatic, gastric, and liver cancers. They also described “atypical” cases that carried only part of the usual whole-genome damage pattern. (aacrjournals.org) The timing tracks a fresh public data release. The National Cancer Institute’s Genomic Data Commons said Data Release 42 added 8,000-plus new whole-genome variant call sets for TCGA and other programs, along with more than 1,500 higher-coverage TCGA whole-genome alignments. (gdc.cancer.gov) TCGA itself had already profiled more than 11,000 tumor-normal pairs across 33 cancer types with multiple molecular methods. What is new here is the scale of whole-genome harmonization on top of that archive, which gives researchers one consistent reference set instead of many separate call files. (aacrjournals.org) The collaboration said it froze the analysis on December 1, 2025 so both groups could compare methods against the same dataset and use it for benchmarking computational and artificial-intelligence models. That kind of fixed reference matters because variant calling can change when pipelines, filters, or alignments change. (businesswire.com) For pathologists and drug developers, the practical shift is from reading cancer genomes as isolated cases to reading them against a much larger atlas of matched tumor-normal genomes. The more complete that atlas gets, the more often small biopsies and cytology samples can be interpreted with whole-genome context rather than exome-era shortcuts. (businesswire.com)