HyperST: Hierarchical Hyperbolic Learning for Spatial Transcriptomics Prediction

2025-11-26

AI Summary

Existing methods for cross-modal prediction of spatial transcriptomics (ST) gene expression from histology images suffer from inadequate spot-level alignment and insufficient hierarchical structure modeling, while failing to address the inherent information asymmetry between visual image representations and molecular gene representations. To overcome these limitations, we propose the first hyperbolic-space-based multi-level image–gene representation learning framework. Our approach introduces a hierarchical hyperspherical alignment mechanism to jointly model the hierarchical semantics of tissue microenvironments (at both spot- and niche-level) and gene expression profiles. By integrating hyperbolic neural networks with hierarchical contrastive learning, we achieve nonlinear cross-modal alignment. Evaluated on four publicly available ST datasets across diverse tissue types, our method achieves state-of-the-art performance, significantly improving gene expression prediction accuracy. This work establishes a novel paradigm for low-cost, high-throughput spatial transcriptomics analysis.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Representation Learning for Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

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

Abstract

Spatial Transcriptomics (ST) merges the benefits of pathology images and gene expression, linking molecular profiles with tissue structure to analyze spot-level function comprehensively. Predicting gene expression from histology images is a cost-effective alternative to expensive ST technologies. However, existing methods mainly focus on spot-level image-to-gene matching but fail to leverage the full hierarchical structure of ST data, especially on the gene expression side, leading to incomplete image-gene alignment. Moreover, a challenge arises from the inherent information asymmetry: gene expression profiles contain more molecular details that may lack salient visual correlates in histological images, demanding a sophisticated representation learning approach to bridge this modality gap. We propose HyperST, a framework for ST prediction that learns multi-level image-gene representations by modeling the data's inherent hierarchy within hyperbolic space, a natural geometric setting for such structures. First, we design a Multi-Level Representation Extractors to capture both spot-level and niche-level representations from each modality, providing context-aware information beyond individual spot-level image-gene pairs. Second, a Hierarchical Hyperbolic Alignment module is introduced to unify these representations, performing spatial alignment while hierarchically structuring image and gene embeddings. This alignment strategy enriches the image representations with molecular semantics, significantly improving cross-modal prediction. HyperST achieves state-of-the-art performance on four public datasets from different tissues, paving the way for more scalable and accurate spatial transcriptomics prediction.

Problem

• Predict gene expression from histology images cost-effectively
• Leverage hierarchical structure of ST data for better alignment
• Bridge modality gap between image and gene expression

Innovation

• Hierarchical hyperbolic alignment for multi-level representation learning
• Multi-level extractors capture spot and niche context from both modalities
• Hyperbolic space modeling bridges image-gene modality gap with molecular semantics