The Cancer Genome Atlas (TCGA) has been instrumental in studying genomic and epigenomic aberrations in a multitude of different human cancers. Corces, Granja et al. now report a rich data resource of accessible chromatin regions for 23 cancer types that provides insight into the landscape of active regulatory DNA elements and transcriptional regulation in human tumours.

Credit: imageBROKER/Alamy

A total of 410 tumour samples were used as input for assay for transposase-accessible chromatin using sequencing (ATAC-seq), which identified more than 500,000 pan-cancer peaks of chromatin accessibility. These peaks overlapped with promoter and enhancer regions defined by chromatin immunoprecipitation followed by sequencing (ChIP–seq) in the ChromHHM database, as expected. Distal elements showed stronger cancer type specificity, whereas proximal, promoter regulatory elements showed similar patterns across cancers.

Based on the genome-wide ATAC-seq patterns, cancer types were clustered into different groups, and cluster-specific peak sets were observed to be enriched for relevant transcription factor (TF) motifs as well as genetic variants identified in genome-wide association studies for these cancers. New cancer subtypes could be defined on the basis of these chromatin accessibility patterns.

Correlating the ATAC-seq peaks with matched RNA sequencing data, the authors were able to predict peak-to-gene links, both across all cancers and specifically for breast cancer. To validate a subset of these links, a CRISPR interference (CRISPRi) approach was used to introduce heterochromatin signatures at select distal peaks. Peaks targeted in this way were associated with a decrease in expression of the linked gene, located kilobases or megabases away. These assays also emphasized the cell type and cancer type specificity of the peak–gene relationships, for example for BCL2 in luminal-like but not basal-like breast cancer.

a rich data resource of accessible chromatin regions for 23 cancer types

Complementary to this unique resource, future studies will be crucial to delineate the contribution of cellular composition of the ATAC-seq signal, to include (matched) healthy tissues and to add additional genomic data such as 3D chromosome conformation to enable researchers to home in on the causative genomic regions in a wide range of human cancers at the full level of complexity of the tumour tissue.