Enhancing the Efficiency of Cell Classification in ScRNA-seq Data by Weakly Supervised Learning

DOI

10.2991/978-94-6239-648-7_62How to use a DOI?

Keywords

Machine Learning; ScRNA-seq; Cell Classification

Abstract

Single-cell RNA sequencing (scRNA-seq) is a genomics technology that enables research of cellular diversity and operation through assessing gene expression. However, no widely available automated classification technology currently exists, and distinguishing cells within scRNA-seq datasets still relies on expert manual annotation. To address that, this study explores weakly supervised learning for automated cell type classification in scRNA-seq. The study applies lightweight machine learning techniques which have been trained with 10% labeled cells using the PBMC 3k dataset. The random forest models and logistic regression were trained on 30 major components and evaluated using their accuracy. Both models were able to classify major immune cell populations, with B cells achieving near-perfect classification. Random Forest performed better in separation of similar subtypes, such as CD4 and CD8 T cells, and exhibited alignment with the clusters in UMAP visualizations. These results indicate that weak supervision may provide efficient results of accurate and meaningful annotations and, therefore, lightweight nonlinear models can be an effective tool in scRNA-seq analysis.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Xinyi Zhou
PY  - 2026
DA  - 2026/04/24
TI  - Enhancing the Efficiency of Cell Classification in ScRNA-seq Data by Weakly Supervised Learning
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 562
EP  - 570
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_62
DO  - 10.2991/978-94-6239-648-7_62
ID  - Zhou2026
ER  -