3D-Guided Scalable Flow Matching for Generating Volumetric Tissue Spatial Transcriptomics from Serial Histology

2025-11-18

AI Summary

This study addresses the challenges of generating high-fidelity, scalable, and structurally coherent three-dimensional (3D) spatial transcriptomic data from serial hematoxylin and eosin (H&E)-stained tissue sections. To this end, we propose HoloTea—a novel generative framework that introduces 3D-guided flow matching for spatial transcriptomic inference. HoloTea integrates a zero-inflated negative binomial prior, cross-section morphological alignment, and ControlNet-based conditional control, while leveraging global attention and a shared feature space to enforce anatomical continuity during gene expression imputation. The framework enables efficient large-scale 3D training and inference across diverse tissues and resolutions. Quantitative and qualitative evaluations demonstrate that HoloTea significantly outperforms state-of-the-art 2D and 3D baselines in accuracy, scalability, and generalizability. It achieves high-fidelity 3D expression prediction, facilitating the reconstruction of volumetric virtual tissues with preserved spatial architecture and molecular fidelity.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: 3D Computer Vision
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for 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
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

A scalable and robust 3D tissue transcriptomics profile can enable a holistic understanding of tissue organization and provide deeper insights into human biology and disease. Most predictive algorithms that infer ST directly from histology treat each section independently and ignore 3D structure, while existing 3D-aware approaches are not generative and do not scale well. We present Holographic Tissue Expression Inpainting and Analysis (HoloTea), a 3D-aware flow-matching framework that imputes spot-level gene expression from H&E while explicitly using information from adjacent sections. Our key idea is to retrieve morphologically corresponding spots on neighboring slides in a shared feature space and fuse this cross section context into a lightweight ControlNet, allowing conditioning to follow anatomical continuity. To better capture the count nature of the data, we introduce a 3D-consistent prior for flow matching that combines a learned zero-inflated negative binomial (ZINB) prior with a spatial-empirical prior constructed from neighboring sections. A global attention block introduces 3D H&E scaling linearly with the number of spots in the slide, enabling training and inference on large 3D ST datasets. Across three spatial transcriptomics datasets spanning different tissue types and resolutions, HoloTea consistently improves 3D expression accuracy and generalization compared to 2D and 3D baselines. We envision HoloTea advancing the creation of accurate 3D virtual tissues, ultimately accelerating biomarker discovery and deepening our understanding of disease.

Problem

• Generating 3D volumetric tissue spatial transcriptomics from histology images
• Overcoming limitations of independent 2D section analysis in transcriptomics prediction
• Enabling scalable 3D-aware gene expression imputation across tissue sections

Innovation

• 3D-aware flow-matching framework for volumetric transcriptomics
• ControlNet conditioning with cross-section anatomical continuity
• Zero-inflated negative binomial prior for count data modeling