76 / 2026-04-05 18:41:50

PanCan3DAtlas: A Pan-cancer 3D Genome Database Revealing Loop-Mediated Regulation and Structural Variation-Driven Chromatin Rewiring

3D genome,Gene regulatory network,Structural variations (SVs),Pan-cancer,multi-omics integrative analysis,Database,Non-coding Variants

Three-dimensional (3D) genome organization plays a key role in transcriptional regulation; however, its systematic characterization across cancer types remains limited. Here, we present PanCan3DAtlas, a comprehensive pan-cancer multi-omics database integrating Hi-C, ATAC-seq, H3K27ac ChIP-seq, and RNA-seq data from tumor and normal samples processed through a unified analytical pipeline, enabling systematic comparison of tumor–normal differences. At the 3D genome level, we performed multi-layered analyses of chromatin compartments, topologically associating domains (TADs), and chromatin loops, identifying cancer type-specific and pan-cancer conserved chromatin features. By integrating multi-omics data to functionally annotate chromatin loops, we constructed enhancer–gene interaction networks that define the 3D regulatory landscape across cancer types.

To demonstrate the utility of PanCan3DAtlas, we used chromatin loops to spatially annotate somatic mutations and identify their potential distal target genes, showing that non-coding mutations can regulate gene expression through 3D genomic interactions. In addition, structural variation (SV) analysis based on Hi-C data was conducted to characterize SV-associated chromatin reorganization, especially loop rearrangements, revealing genes recurrently affected by structural alterations across cancers.

All data and analytical results are integrated into an online platform, enabling interactive visualization and querying of pan-cancer multi-layered 3D genomic information. The platform provides three main functions: (1) gene and genomic region annotation, allowing users to explore 3D interaction networks in specific cancers or across pan-cancer; for gene queries, the platform identifies whether the queried gene functions as a hub within the tumor 3D regulatory network, and displays tumor–normal differences in gene expression and epigenomic signals across cancer types; (2) structural variation module, presenting SV events and their associated target genes affected by 3D chromatin remodeling; (3) sample-level exploration, providing integrated visualization of Hi-C contact maps, epigenomic signal tracks, and gene expression data.

In summary, PanCan3DAtlas integrates multi-omics and 3D genomic data into a unified and accessible resource, enabling systematic investigation of spatial chromatin organization, structural variation, and their regulatory impacts across cancers, and providing a valuable platform for pan-cancer 3D genome research and functional interpretation of non-coding variants.