scLLM-DSC: LLM-Knowledge Enhanced Cross-Modal Deep Structural Clustering for Single-Cell RNA Sequencing

2026-06-11

AI Summary

This study addresses a critical limitation in existing single-cell RNA sequencing clustering methods, which predominantly rely on numerical statistical patterns while neglecting the biological semantics of genes, thereby creating semantic blind spots. To overcome this, the authors propose a novel framework that uniquely integrates biological prior knowledge from large language models with graph-guided encoding to construct a knowledge-driven semantic view and a structure-aware topological view. These dual representations are jointly modeled within a unified latent space through a cross-modal contrastive alignment mechanism, enabling deep structural clustering that effectively bridges the inherent gap between generative pretraining and discriminative clustering objectives. Extensive benchmark evaluations demonstrate that the proposed method significantly outperforms eleven state-of-the-art approaches, achieving substantial improvements in clustering accuracy.

Abstract

Clustering is fundamental to scRNA-seq analysis, serving as a cornerstone for identifying cell populations and resolving tissue heterogeneity. However, existing methods focus on mining numerical statistical patterns, suffering from semantic agnosticism by neglecting the intrinsic biological functions encoded by genes. While Large Language Models (LLMs) offer promising semantic capabilities, their direct adaptation to cell clustering is hindered by the structural mismatch between generative pre-training objectives and discriminative downstream tasks. To bridge this gap, we propose scLLM-DSC, a novel LLM-Knowledge Enhanced Cross-Modal Deep Structural Clustering framework. Diverging from data-driven paradigms, scLLM-DSC establishes a semantically-grounded representation by synergizing two views: a Knowledge-Driven Semantic View derived from NCBI gene priors and contextualized Cell2Sentence embeddings, and a Structure-Aware Topological View extracted via a graph-guided encoder. Crucially, we introduce a cross-modal contrastive alignment mechanism to enforce consistency between biological semantics and transcriptomic features within a unified latent space. Extensive benchmarks demonstrate that scLLM-DSC significantly outperforms eleven state-of-the-art baselines in clustering accuracy.

Problem

• single-cell RNA sequencing
• clustering
• semantic agnosticism
• Large Language Models
• biological function

Innovation

• LLM-knowledge integration
• cross-modal contrastive alignment
• semantic-aware clustering
• graph-guided encoder
• single-cell RNA sequencing