When Genes Speak: A Semantic-Guided Framework for Spatially Resolved Transcriptomics Data Clustering

2025-11-14

AI Summary

Current spatial transcriptomics analysis models treat genes as isolated numerical features, neglecting the biologically meaningful semantics encoded in gene symbols—thereby limiting functional interpretability. To address this, we propose SemST, a semantic-guided deep clustering framework that pioneers the integration of large language model (LLM)-derived gene semantic embeddings with graph neural network (GNN)-based modeling of spatial neighborhood structures. Its core innovation is the Fine-grained Semantic Modulation (FSM) module, which performs element-wise, dynamic calibration of spatial features via learnable affine transformations conditioned on gene semantics, offering plug-and-play compatibility. Evaluated on multiple public spatial transcriptomics datasets, SemST consistently outperforms state-of-the-art methods. Ablation studies confirm that FSM robustly enhances baseline performance across diverse settings, empirically validating the efficacy and generalizability of semantic guidance in spatial transcriptomics analysis.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

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: Large language models for search

Abstract

Spatial transcriptomics enables gene expression profiling with spatial context, offering unprecedented insights into the tissue microenvironment. However, most computational models treat genes as isolated numerical features, ignoring the rich biological semantics encoded in their symbols. This prevents a truly deep understanding of critical biological characteristics. To overcome this limitation, we present SemST, a semantic-guided deep learning framework for spatial transcriptomics data clustering. SemST leverages Large Language Models (LLMs) to enable genes to"speak"through their symbolic meanings, transforming gene sets within each tissue spot into biologically informed embeddings. These embeddings are then fused with the spatial neighborhood relationships captured by Graph Neural Networks (GNNs), achieving a coherent integration of biological function and spatial structure. We further introduce the Fine-grained Semantic Modulation (FSM) module to optimally exploit these biological priors. The FSM module learns spot-specific affine transformations that empower the semantic embeddings to perform an element-wise calibration of the spatial features, thus dynamically injecting high-order biological knowledge into the spatial context. Extensive experiments on public spatial transcriptomics datasets show that SemST achieves state-of-the-art clustering performance. Crucially, the FSM module exhibits plug-and-play versatility, consistently improving the performance when integrated into other baseline methods.

Problem

• Integrating gene biological semantics with spatial transcriptomics data clustering
• Overcoming limitations of treating genes as isolated numerical features
• Developing framework to combine symbolic gene meanings with spatial relationships

Innovation

• LLMs transform gene symbols into biological embeddings
• GNNs capture spatial neighborhood relationships for integration
• FSM module calibrates spatial features with semantic embeddings