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

TL;DR

HaDM-ST is a novel framework that leverages histology images and low-resolution spatial transcriptomics data to generate high-resolution, gene-specific spatial gene expression maps with improved accuracy and fidelity.

Contribution

It introduces a comprehensive model combining semantic feature extraction, precise spatial alignment, and gene-specific adversarial learning for enhanced ST resolution.

Findings

Outperforms prior methods in spatial fidelity and gene coherence.

Effective across diverse tissues and species.

Enhances high-resolution spatial transcriptomics generation.

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.