A genome-wide single-cell 3D genome atlas of lung cancer progression

Alterations in three-dimensional (3D) genome structures are associated with cancer[1][1]–[5][2]. However, how genome folding evolves and diversifies during subclonal cancer progression in the native tissue environment remains unknown. Here, we leveraged a genome-wide chromatin tracing technology to directly visualize 3D genome folding in situ in a faithful Kras -driven mouse model of lung adenocarcinoma (LUAD)[6][3], generating the first single-cell 3D genome atlas of any cancer. We discovered stereotypical 3D genome alterations during cancer development, including a striking structural bottleneck in preinvasive adenomas prior to progression to LUAD, indicating a stringent selection on the 3D genome early in cancer progression. We further showed that the 3D genome precisely encodes cancer states in single cells, despite considerable cell-to-cell heterogeneity. Finally, evolutionary changes in 3D genome compartmentalization – partially regulated by polycomb group protein Rnf2 through its ubiquitin ligase-independent activity – reveal novel genetic drivers and suppressors of LUAD progression. Our results demonstrate the importance of mapping the single-cell cancer 3D genome and the potential to identify new diagnostic and therapeutic biomarkers from 3D genomic architectures. ### Competing Interest Statement S.W., M.L., M.D.M., S.J., and S.S.A. are inventors on a patent applied for by Yale University related to this work. M.D.M. received research funding from a Genentech supported AACR grant. S.W. is a share-holder and consultant of Translura, Inc. The remaining authors declare no competing interests. [1]: #ref-1 [2]: #ref-5 [3]: #ref-6