Amazing news KJ!!! Very well deserved! Congratulations!!!🎉 🎉🎉

Congratulations Christa!!! Wishing you many productive and fun lab decades ahead! That seems delicious!!!

Special thanks to the @themarkfdn.bsky.social for believing and supporting this project from the very beginning and to NIH for their funding! #3DGenome #Glioblastoma

Congratulations to @SarahBreves, @dafnecdg.bsky.social and Alex Polyzos for driving this study and many thanks to the Fine group and all collaborators for their critical contributions!!! Grateful also to Howard Chang for sharing their data.

What drives 3D hub formation? Intriguingly, only a small fraction of 3D hubs overlapped with predicted structural variants, such as amplifications, suggesting epigenetic mechanisms. Transcription factors with enriched binding on hubs were often predictive of worse outcomes.

What about other cancers? Using published HiChIP datasets across 16 cancer types we also identified shared and tumor-specific 3D hubs associated with universal or cancer-specific oncogenic programs.

CRISPRi targeting of a highly recurrent hub -with previously unknown role in GBM- was sufficient to cause concordant downregulation of multiple hub-connected genes, global transcriptional shifts and reduced clonogenicity.

Moreover, gene pairs within hubs showed significantly higher coregulation in single-cells compared to genes within the same TAD or in proximity. However, for hubs connected to multiple genes the mean coregulation is highly variable likely reflecting intratumoral heterogeneity.

By mapping 3D enhancer-promoter networks in patient-derived glioblastoma stem cells, we identified hyperconnected 3D hubs that strongly enriched for universal and tumor-specific programs and accurately distinguished aggressive IDH-wt from lower-grade glioma patients.