scDCCA: deep contrastive clustering for single-cell RNA-seq data based on auto-encoder network

• Published in: Briefings in bioinformatics Vol. 24; no. 1
• Main Authors: Wang, Jing, Xia, Junfeng, Wang, Haiyun, Su, Yansen, Zheng, Chun-Hou
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 19.01.2023; Oxford Publishing Limited (England)
• Subjects: Algorithms; Biological analysis; Cluster Analysis; Clustering; Coders; Gene Expression Profiling - methods; Gene sequencing; Heterogeneity; Humans; Learning; Machine learning; Modules; Noise reduction; Representations; Sequence Analysis, RNA - methods; Single-Cell Analysis - methods; Single-Cell Gene Expression Analysis; contrastive learning; scRNA-seq data; deep clustering; denoising auto-encoder; ZINB model
• ISSN: 1467-5463; 1477-4054; 1477-4054
• DOI: 10.1093/bib/bbac625

Abstract The advances in single-cell ribonucleic acid sequencing (scRNA-seq) allow researchers to explore cellular heterogeneity and human diseases at cell resolution. Cell clustering is a prerequisite in scRNA-seq analysis since it can recognize cell identities. However, the high dimensionality, noises and significant sparsity of scRNA-seq data have made it a big challenge. Although many methods have emerged, they still fail to fully explore the intrinsic properties of cells and the relationship among cells, which seriously affects the downstream clustering performance. Here, we propose a new deep contrastive clustering algorithm called scDCCA. It integrates a denoising auto-encoder and a dual contrastive learning module into a deep clustering framework to extract valuable features and realize cell clustering. Specifically, to better characterize and learn data representations robustly, scDCCA utilizes a denoising Zero-Inflated Negative Binomial model-based auto-encoder to extract low-dimensional features. Meanwhile, scDCCA incorporates a dual contrastive learning module to capture the pairwise proximity of cells. By increasing the similarities between positive pairs and the differences between negative ones, the contrasts at both the instance and the cluster level help the model learn more discriminative features and achieve better cell segregation. Furthermore, scDCCA joins feature learning with clustering, which realizes representation learning and cell clustering in an end-to-end manner. Experimental results of 14 real datasets validate that scDCCA outperforms eight state-of-the-art methods in terms of accuracy, generalizability, scalability and efficiency. Cell visualization and biological analysis demonstrate that scDCCA significantly improves clustering and facilitates downstream analysis for scRNA-seq data. The code is available at https://github.com/WJ319/scDCCA.