scCDCG: Efficient Deep Structural Clustering for single-cell RNA-seq via Deep Cut-informed Graph Embedding

2024-04-09

International Conference on Database Systems for Advanced Applications

AI Summary

To address insufficient structural information utilization, graph neural network (GNN) over-smoothing, and computational inefficiency caused by high dimensionality and sparsity in single-cell RNA-seq (scRNA-seq) clustering, this paper proposes DeepCut—a novel clustering framework integrating deep cut-aware graph embedding, optimal transport-guided self-supervised learning, and a lightweight variational autoencoder (VAE). DeepCut introduces the first deep cut optimization mechanism for graph embedding to mitigate GNN over-smoothing; designs an optimal transport-based self-supervised paradigm tailored to scRNA-seq’s high-dimensional sparsity; and enables synergistic, efficient modeling of cellular expression structure via its three co-designed modules. Evaluated on six real-world scRNA-seq datasets, DeepCut achieves an average 5.2% improvement in clustering accuracy over seven state-of-the-art methods and runs 3.8× faster. The implementation is publicly available.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Learning & Optimization for NLP
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Single-cell RNA sequencing (scRNA-seq) is essential for unraveling cellular heterogeneity and diversity, offering invaluable insights for bioinformatics advancements. Despite its potential, traditional clustering methods in scRNA-seq data analysis often neglect the structural information embedded in gene expression profiles, crucial for understanding cellular correlations and dependencies. Existing strategies, including graph neural networks, face challenges in handling the inefficiency due to scRNA-seq data's intrinsic high-dimension and high-sparsity. Addressing these limitations, we introduce scCDCG (single-cell RNA-seq Clustering via Deep Cut-informed Graph), a novel framework designed for efficient and accurate clustering of scRNA-seq data that simultaneously utilizes intercellular high-order structural information. scCDCG comprises three main components: (i) A graph embedding module utilizing deep cut-informed techniques, which effectively captures intercellular high-order structural information, overcoming the over-smoothing and inefficiency issues prevalent in prior graph neural network methods. (ii) A self-supervised learning module guided by optimal transport, tailored to accommodate the unique complexities of scRNA-seq data, specifically its high-dimension and high-sparsity. (iii) An autoencoder-based feature learning module that simplifies model complexity through effective dimension reduction and feature extraction. Our extensive experiments on 6 datasets demonstrate scCDCG's superior performance and efficiency compared to 7 established models, underscoring scCDCG's potential as a transformative tool in scRNA-seq data analysis. Our code is available at: https://github.com/XPgogogo/scCDCG.

Problem

• Captures intercellular high-order structural information from scRNA-seq data
• Overcomes inefficiency due to high-dimension and high-sparsity of scRNA-seq
• Provides efficient and accurate clustering through deep cut-informed graph embedding

Innovation

• Deep cut-informed graph embedding captures intercellular structural information
• Optimal transport guides self-supervised learning for sparse data
• Autoencoder reduces dimensions and extracts features efficiently