Accurate prediction of ecDNA in interphase cancer cells using deep neural networks
Oncogene amplification is a key driver of cancer pathogenesis and is often mediated by extrachromosomal DNA (ecDNA). EcDNA amplifications are associated with increased pathogenicity of cancer and poorer outcomes for patients. EcDNA can be detected accurately using fluorescence in situ hybridization (FISH) when cells are arrested in metaphase. However, the majority of cancer cells are non-mitotic and must be analyzed in interphase, where it is difficult to discern extrachromosomal amplifications from chromosomal amplifications. Thus, there is a need for methods that accurately predict oncogene amplification status from interphase cells.We present interSeg, a deep learning-based tool to cytogenetically classify oncogene amplification status as extrachromosomally amplified (EC-amp), intrachromosomally amplified (HSR-amp), or not amplified, from interphase FISH images. We trained and validated interSeg on 652 images (40,446 nuclei). Tests on 215 cultured cell and tissue model images (9,733 nuclei) showed 89% and 97% accuracy at the nuclear and sample levels, respectively. The neuroblastoma patient tissue hold-out set (67 samples and 1,937 nuclei) also revealed 97% accuracy at the sample level in detecting the presence of focal amplification. In experimentally and computationally mixed images, interSeg accurately predicted the level of heterogeneity. The results showcase interSeg as an important method for analyzing oncogene amplifications.
Authors
- Anton George Henssen (ORCID: https://orcid.org/0000-0003-1534-778X)
- Sihan Wu (ORCID: https://orcid.org/0000-0001-8329-7492)
- Gino Prasad (ORCID: https://orcid.org/0000-0003-4590-1278)
- Vineet Bafna (ORCID: https://orcid.org/0000-0002-5810-6241)
- Xicheng Zhang (ORCID: https://orcid.org/0000-0003-1004-021X)
- Ellis J. Curtis (ORCID: https://orcid.org/0000-0003-3751-9968)
- Howard Y. Chang (ORCID: https://orcid.org/0000-0002-9459-4393)
- Paul S. Mischel (ORCID: https://orcid.org/0000-0002-4560-2211)
- Julie Wiese
- Lotte Brückner (ORCID: https://orcid.org/0000-0002-6155-2277)
- Kristen M. Turner (ORCID: https://orcid.org/0000-0002-4661-3094)
- Ivy Tsz-Lo Wong (ORCID: https://orcid.org/0000-0002-9761-3422)
- Homa Hemmati
- Matthias Fischer (ORCID: https://orcid.org/0000-0003-1363-1242)
- Utkrisht Rajkumar (ORCID: https://orcid.org/0000-0002-2825-6310)
- Justin Wahl
- Jessica Theissen
- Xiaowei Yan (ORCID: https://orcid.org/0000-0002-4846-8812)
Institutions
- Amgen (United States) (US)
- Howard Hughes Medical Institute (US)
- University of Cologne (DE)
- Max Delbrück Center (DE)
- University of California San Diego (US)
- Boundless Bio (United States) (US)
- University Hospital Cologne (DE)
- Charité - Universitätsmedizin Berlin (DE)
- The University of Texas Southwestern Medical Center (US)
- Stanford University (US)
Publication Details
- Journal
- Communications Biology
- Published
- 2026-04-11
- DOI
- https://doi.org/10.1038/s42003-026-09982-4
- Primary Topic
- Cancer Genomics and Diagnostics
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Science Foundation
- U.S. Department of Health and Human Services
- Damon Runyon Cancer Research Foundation
- Cancer Research UK
- National Cancer Institute
- National Institute of General Medical Sciences