Refinement Contrastive Learning of Cell-Gene Associations for Unsupervised Cell Type Identification

TL;DR

This paper introduces scRCL, a novel contrastive learning framework that explicitly models cell-gene associations to improve unsupervised cell type identification in single-cell omics data.

Contribution

The method uniquely integrates cell-gene interactions into contrastive learning, enhancing cell embedding quality for better clustering and biological interpretability.

Findings

Outperforms state-of-the-art methods in cell-type identification accuracy

Recovers biologically meaningful gene-expression signatures

Demonstrates robustness across multiple datasets

Abstract

Unsupervised cell type identification is crucial for uncovering and characterizing heterogeneous populations in single cell omics studies. Although a range of clustering methods have been developed, most focus exclusively on intrinsic cellular structure and ignore the pivotal role of cell-gene associations, which limits their ability to distinguish closely related cell types. To this end, we propose a Refinement Contrastive Learning framework (scRCL) that explicitly incorporates cell-gene interactions to derive more informative representations. Specifically, we introduce two contrastive distribution alignment components that reveal reliable intrinsic cellular structures by effectively exploiting cell-cell structural relationships. Additionally, we develop a refinement module that integrates gene-correlation structure learning to enhance cell embeddings by capturing underlying cell-gene associations. This module strengthens connections between cells and their associated genes, refining the representation learning to exploiting biologically meaningful relationships. Extensive experiments on several single-cell RNA-seq and spatial transcriptomics benchmark datasets demonstrate that our method consistently outperforms state-of-the-art baselines in cell-type identification accuracy. Moreover, downstream biological analyses confirm that the recovered cell populations exhibit coherent gene-expression signatures, further validating the biological relevance of our approach. The code is available at https://github.com/THPengL/scRCL.