Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 5 min ago
PreprintCC BYbioRxiv

scOLAR: Ontology-Anchored Open-Set Annotation of Single-Cell RNA-seq Data

  • Y. Liu
  • S. Yi
  • H. Yin
  • W. Ju1
  1. 1School of Artificial Intelligence, Sichuan University
Read full textPDFDOICite

Abstract

Single-cell RNA sequencing profiles cellular heterogeneity at atlas scale, making automated annotation essential. However, target datasets often contain novel cell types missing from incomplete references. We present scOLAR, an ontology-guided open-set framework that learns prototypes over the Cell Ontology and uses both reference and target expression to annotate known classes while detecting unfamiliar populations. Guided by ontology hierarchies and decision-boundary regularization, scOLAR penalizes coarse-lineage misclassification and groups novel cells without requiring predefined cluster counts. Across benchmarks, scOLAR achieves a novelty-detection AUROC of 0.9726 and an average precision of 0.9871, enabling structured post-hoc lineage-level interpretation of populations absent from the reference.

Related

  1. GoCo: structure-guided conformal calibration for error-controlled Gene Ontology annotation release
    Chong Wang, Yongzhao Shao, Peng Liu
    arXivPreprintarXiv licenceFull textBioinformatics & computational biologyStatisticsarXiv
  2. GOParsimony: parsimonious saltation across the Gene Ontology digraph for efficient functional annotation
    P. M. Harrison
    bioRxivPreprintCC BY-NC-NDFull textBioinformatics & computational biologybioRxiv
  3. Ontology-Driven Annotation of Key Events Enables Transcriptomics-Based Assessment of Adverse Outcome Pathway Networks
    S. Adil, T. M. de Kok, E. Willighagen +2
    bioRxivPreprintCC BY-NC-NDFull textBioinformatics & computational biologybioRxiv
  4. AVOCODO: An open-source multimodal annotation platform for developmental EEG
    W. W. An
    bioRxivPreprintCC BY-NC-NDFull textBioinformatics & computational biologybioRxiv
  5. Where single-cell foundation models help: a benchmark of annotation, perturbation and regulatory inference against simple baselines
    F. A. Cupaioli, I. Merelli
    bioRxivPreprintAll rights reservedFull textBioinformatics & computational biologybioRxiv

Cite

BibTeXRIS
BibTeX
@misc{liu2026scolar,
  title = {{scOLAR: Ontology-Anchored Open-Set Annotation of Single-Cell RNA-seq Data}},
  author = {Y. Liu and S. Yi and H. Yin and W. Ju},
  howpublished = {bioRxiv},
  year = {2026},
  month = sep,
  doi = {10.64898/2026.09.05.749179},
  url = {https://www.biorxiv.org/content/10.64898/2026.09.05.749179v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - scOLAR: Ontology-Anchored Open-Set Annotation of Single-Cell RNA-seq Data
AU  - Y. Liu
AU  - S. Yi
AU  - H. Yin
AU  - W. Ju
PY  - 2026
DA  - 2026/09/05
PB  - bioRxiv
DO  - 10.64898/2026.09.05.749179
UR  - https://www.biorxiv.org/content/10.64898/2026.09.05.749179v1
AB  - Single-cell RNA sequencing profiles cellular heterogeneity at atlas scale, making automated annotation essential. However, target datasets often contain novel cell types missing from incomplete references. We present scOLAR, an ontology-guided open-set framework that learns prototypes over the Cell Ontology and uses both reference and target expression to annotate known classes while detecting unfamiliar populations. Guided by ontology hierarchies and decision-boundary regularization, scOLAR penalizes coarse-lineage misclassification and groups novel cells without requiring predefined cluster counts. Across benchmarks, scOLAR achieves a novelty-detection AUROC of 0.9726 and an average precision of 0.9871, enabling structured post-hoc lineage-level interpretation of populations absent from the reference.
ER  -