HaDM-ST: Histology-Assisted Differential Modeling for Spatial Transcriptomics Generation

2025-08-10

AI Summary

Current spatial transcriptomics (ST) technologies suffer from limited platform resolution, hindering accurate characterization of spatial heterogeneity in gene expression. Existing H&E image–based super-resolution methods face three key challenges: difficulty in extracting expression-relevant histological features from H&E, imprecise pixel-level cross-modal alignment between H&E and ST, and insufficient modeling of gene-specific spatial variation. To address these, we propose DiffST—a diffusion model–based framework for generating high-resolution ST data. Its core innovations include: (1) a semantic distillation network that disentangles expression-correlated histological features from H&E; (2) a spatial alignment module enabling precise pixel-level cross-modal registration between H&E and ST; and (3) channel-aware adversarial learning to capture gene-specific spatial variability. Evaluated on 200 genes across multiple tissues and species, DiffST significantly improves both spatial fidelity and gene expression consistency, outperforming state-of-the-art methods.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Spatial transcriptomics (ST) reveals spatial heterogeneity of gene expression, yet its resolution is limited by current platforms. Recent methods enhance resolution via H&E-stained histology, but three major challenges persist: (1) isolating expression-relevant features from visually complex H&E images; (2) achieving spatially precise multimodal alignment in diffusion-based frameworks; and (3) modeling gene-specific variation across expression channels. We propose HaDM-ST (Histology-assisted Differential Modeling for ST Generation), a high-resolution ST generation framework conditioned on H&E images and low-resolution ST. HaDM-ST includes: (i) a semantic distillation network to extract predictive cues from H&E; (ii) a spatial alignment module enforcing pixel-wise correspondence with low-resolution ST; and (iii) a channel-aware adversarial learner for fine-grained gene-level modeling. Experiments on 200 genes across diverse tissues and species show HaDM-ST consistently outperforms prior methods, enhancing spatial fidelity and gene-level coherence in high-resolution ST predictions.

Problem

• Extracting expression-relevant features from complex H&E images
• Achieving precise multimodal alignment in diffusion frameworks
• Modeling gene-specific variation across expression channels

Innovation

• Semantic distillation network extracts H&E predictive cues
• Spatial alignment module ensures pixel-wise ST correspondence
• Channel-aware adversarial learner models gene-specific variation