Peer-Reviewed Publication
bioRxiv2026July 11, 2026Journal Article
Scalable 3D cell-interaction analysis via supercell graphs for prostate cancer risk stratification.
Yujie Zhao1,2, Sarah S L Chow2,3, Renao Yan2,3, David Brenes2, Robert Serafin4, Cristina Almagro-Pérez5,6,7,8, Andrew H Song9,10, Priti Lal11, Emily Chan12, Michelle Downes13, Elena Baraznenok14, Jennifer Salguero Lopez15, Anant Madabhush15, Faisal Mahmood5,6,7,16, Lawrence D True17, Jonathan T C Liu2,3,17,18
1Department of Biomedical Informatics and Medical Education, University of Washington, Seattle, WA, USA.
2Department of Pathology, Stanford University, Stanford, CA, USA.
3Department of Mechanical Engineering, University of Washington, Seattle, WA, USA.
4Department of Medicine, Section of Hematology/Oncology, University of Chicago, Chicago, IL, USA.
5Department of Pathology, Mass General Brigham, Harvard Medical School, Boston, MA, USA.
6Cancer Program, Broad Institute of Harvard and MIT, Cambridge, MA, USA.
7Data Science Program, Dana-Farber Cancer Institute, Boston, MA, USA.
8Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, MA, USA.
9Department of Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
10Institute for Data Science in Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
11Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USA.
12Stanford Medical Center, Stanford, CA, USA.
13Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada.
14UC Berkeley-UCSF Graduate Program in Bioengineering, University of California, Berkeley, CA, USA.
15Wallace H. Coulter Department of Biomedical Engineering, Emory University & Georgia Institute of Technology, Atlanta, GA, USA.
16Harvard Data Science Initiative, Harvard University, Cambridge, MA, USA.
17Department of Laboratory Medicine & Pathology, University of Washington, Seattle, WA, USA.
18Department of Bioengineering, Stanford University, Stanford, CA, USA.
Abstract
Cellular interactions underlie fundamental biological processes but are not fully represented in conventional 2D histology images. While 3D pathology allows for more-accurate construction of cell-level graphs, machine-learning models are computationally unwieldy and prone to overfitting, especially when dealing with small cohorts. Here, we introduce SCALE3D, a SuperCell graph Analysis framework fo…
Create a free account to keep reading
Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.
Want unlimited research access? See Pro plans