Clustering by Denoising: Latent plug-and-play diffusion for single-cell data

2025-10-26

AI Summary

In single-cell RNA sequencing (scRNA-seq), measurement noise and biological variability cause cell type overlap in latent space, limiting clustering accuracy. To address this, we propose DecDiff—a plug-and-play diffusion framework that decouples the observation space from a denoised latent space. Its core innovation is an input-space-guided Gibbs sampling scheme: it imposes a diffusion prior in a low-dimensional latent space while enabling structure-preserving, adaptive denoising in the high-dimensional original space. DecDiff supports uncertainty quantification and leverages clean reference data to enhance generalization. Evaluated on synthetic benchmarks and diverse real-world scRNA-seq datasets, DecDiff significantly improves clustering accuracy. Cluster boundaries align more closely with known marker genes and developmental trajectories, and biological consistency—assessed via functional enrichment and trajectory coherence—is markedly enhanced.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Single-cell RNA sequencing (scRNA-seq) enables the study of cellular heterogeneity. Yet, clustering accuracy, and with it downstream analyses based on cell labels, remain challenging due to measurement noise and biological variability. In standard latent spaces (e.g., obtained through PCA), data from different cell types can be projected close together, making accurate clustering difficult. We introduce a latent plug-and-play diffusion framework that separates the observation and denoising space. This separation is operationalized through a novel Gibbs sampling procedure: the learned diffusion prior is applied in a low-dimensional latent space to perform denoising, while to steer this process, noise is reintroduced into the original high-dimensional observation space. This unique "input-space steering" ensures the denoising trajectory remains faithful to the original data structure. Our approach offers three key advantages: (1) adaptive noise handling via a tunable balance between prior and observed data; (2) uncertainty quantification through principled uncertainty estimates for downstream analysis; and (3) generalizable denoising by leveraging clean reference data to denoise noisier datasets, and via averaging, improve quality beyond the training set. We evaluate robustness on both synthetic and real single-cell genomics data. Our method improves clustering accuracy on synthetic data across varied noise levels and dataset shifts. On real-world single-cell data, our method demonstrates improved biological coherence in the resulting cell clusters, with cluster boundaries that better align with known cell type markers and developmental trajectories.

Problem

• Improving single-cell RNA-seq clustering accuracy by handling measurement noise
• Separating observation and denoising spaces using diffusion framework
• Enhancing biological coherence of cell clusters through adaptive denoising

Innovation

• Latent plug-and-play diffusion separates observation and denoising spaces
• Gibbs sampling applies diffusion prior in low-dimensional latent space
• Input-space steering maintains denoising fidelity to original data structure