SCUDDO: an unsupervised clustering algorithm for single-cell Hi-C maps using diagonal diffusion operators
Bioinformatics · 03.05.2026 · 42 (5)
DOI: 10.1093/bioinformatics/btag284 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 11.05.2026. Verfasst von Luka Maisuradze und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT 06520, United States. Hauptthemen laut MEDLINE: Single-Cell Analysis, Chromatin, Algorithms, Software. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.
Abstract (Originalsprache)
Motivation Advances in high-throughput chromatin conformation capture have provided insight into the three-dimensional structure and organization of chromatin. While bulk Hi-C experiments capture spatio-temporally averaged chromatin interactions across millions of cells, single-cell Hi-C experiments report on the chromatin interactions of individual cells. Supervised and unsupervised algorithms have been developed to embed single-cell Hi-C maps and identify different cell types. However, single-cell Hi-C maps are often difficult to cluster due to their high sparsity, with state-of-the-art algorithms achieving a maximum Adjusted Rand Index (ARI) of only ≲0.4 on several datasets. Results We introduce a novel unsupervised algorithm, Single-cell Clustering Using Diagonal Diffusion Operators (SCUDDO), to embed and cluster single-cell Hi-C maps. We evaluate SCUDDO on four previously difficult-to-cluster single-cell Hi-C datasets, and show that it can outperform other current algorithms in ARI by ≳0.2. Further, SCUDDO outperforms all other tested algorithms even when we restrict the number of intrachromosomal maps for each cell type and when we use only a small fraction of contacts in each Hi-C map. Thus, SCUDDO can capture the underlying latent features of single-cell Hi-C maps and provide accurate labelling of cell types even when cell types are not known a priori. Availability and implementation SCUDDO is freely available at https://www.github.com/lmaisuradze/scuddo as well as https://doi.org/10.6084/m9.figshare.31759915. The tested datasets are publicly available and can be downloaded from the Gene Expression Omnibus.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT 06520, United States.
- Benjamin Levich Institute and Physics Department, The City College of New York, New York, NY 10031, United States.
- Department of Mechanical Engineering, Yale University, New Haven, CT 06520, United States.
- Graduate Program in Computational Biology and Biomedical Informatics, Yale University, New Haven, CT 06520, United States.
ORCID-Profile: Luka Maisuradze, Corey S O'Hern
Förderung
- National Science Foundation
Zitieren
APA
Maisuradze, L., Shattuck, M. D., & O'Hern, C. S. (2026). SCUDDO: an unsupervised clustering algorithm for single-cell Hi-C maps using diagonal diffusion operators. Bioinformatics, 42(5). https://doi.org/10.1093/bioinformatics/btag284
Vancouver
Maisuradze L, Shattuck MD, O'Hern CS. SCUDDO: an unsupervised clustering algorithm for single-cell Hi-C maps using diagonal diffusion operators. Bioinformatics. 2026;42(5). doi:10.1093/bioinformatics/btag284
BibTeX
@article{Maisuradze2026,
author = {Maisuradze, Luka and Shattuck, Mark D and O'Hern, Corey S},
title = {SCUDDO: an unsupervised clustering algorithm for single-cell Hi-C maps using diagonal diffusion operators},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {5},
doi = {10.1093/bioinformatics/btag284}
}
EUVIMED