Scalable Generation of Spatial Transcriptomics from Histology Images via Whole-Slide Flow Matching
Tinglin Huang, Tianyu Liu, Mehrtash Babadi, Wengong Jin, Rex Ying
Abstract
Spatial transcriptomics (ST) has emerged as a powerful technology for bridging histology imaging with gene expression profiling. However, its application has been limited by low throughput and the need for specialized experimental facilities. Prior works sought to predict ST from wholeslide histology images to accelerate this process, but they suffer from two major limitations. First, they do not explicitly model cell-cell interaction as they factorize the joint distribution of wholeslide ST data and predict the gene expression of each spot independently. Second, their encoders struggle with memory constraints due to the large number of spots (often exceeding 10,000) in typical ST datasets. Herein, we propose STFlow, a flow matching generative model that considers cell-cell interaction by modeling the joint distribution of gene expression of an entire slide. It also employs an efficient slide-level encoder with local spatial attention, enabling whole-slide processing without excessive memory overhead. On the recently curated HEST-1k and STImage-1K4M benchmarks, STFlow substantially outperforms state-of-the-art baselines and achieves over 18% relative improvements over the pathology foundation models.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
9 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Frame Averaging for Invariant and Equivariant Network Design cites GemNet: Universal Directional Graph Neural Networks for Molecules
- Frame Averaging for Invariant and Equivariant Network Design cites E(n) Equivariant Graph Neural Networks
Cited by top-tier papers9
- CARE: A Molecular-Guided Foundation Model with Adaptive Region Modeling for Whole Slide Image AnalysisDi Zhang, Zhangpeng Gong, Xiaobo Pang, Jiashuai Liu et al.CVPR 2026 · 11 citations
- HistoPrism: Unlocking Functional Pathway Analysis from Pan-Cancer Histology via Gene Expression PredictionSusu Hu, Qinghe Zeng, Nithya Bhasker, Jakob Nikolas Kather et al.ICLR 2026 · 5 citations
- Adapting a Pre-trained Single-Cell Foundation Model to Spatial Gene Expression Generation from Histology ImagesDonghai Fang, Yongheng Li, Zhen WANG, Yuansong Zeng et al.CVPR 2026 · 2 citations
- SPATIA: Multimodal Generation and Prediction of Spatial Cell PhenotypesZhenglun Kong, Mufan Qiu, John Boesen, xiang lin et al.ICML 2026 · 1 citation
- FLAG: Foundation model representation with Latent diffusion Alignment via Graph for spatial gene expression predictionQi Si, Penglei Wang, Yushuai Wu, Yifeng Jiao et al.ICML 2026 · 1 citation
Builds on13
- How Attentive are Graph Attention Networks?Shaked Brody, Uri Alon, Eran YahavICLR 2022 · 1,717 citations
- E(n) Equivariant Graph Neural NetworksVictor Garcia Satorras, Emiel Hoogeboom, Max WellingICML 2021 · 1,432 citations
- GemNet: Universal Directional Graph Neural Networks for MoleculesJohannes Gasteiger, Florian Becker, Stephan GünnemannNeurIPS 2021 · 665 citations
- AlphaFold Meets Flow Matching for Generating Protein EnsemblesBowen Jing, Bonnie Berger, Tommi S. JaakkolaICML 2024 · 229 citations
- Frame Averaging for Invariant and Equivariant Network DesignOmri Puny, Matan Atzmon, Edward J. Smith, Ishan Misra et al.ICLR 2022 · 177 citations
Related papers
- Predicting Spatial Transcriptomics from Histology Images via High-Order Multi-Cell Interaction ModelingYouhan Sun, Jiahua Rao, Kangrui Du, Jiancong Xie et al.CVPR 2026
- HiFusion: Hierarchical Intra-Spot Alignment and Regional Context Fusion for Spatial Gene Expression Prediction from HistopathologyZiqiao Weng, Yaoyu Fang, Jiahe Qian, Xinkun Wang et al.AAAI 2026
- RNA-FM: Flow-Matching Generative Model for Genome-wide RNA-Seq PredictionYaxuan Song, Jianan Fan, Tianyi Wang, Qiuyue Hu et al.ICML 2026
- HiST: A Hierarchical Sparse Transformer for Cross-Modal Spatial Transcriptomics ModelingWeiyi Wu, Xinwen Xu, Xingjian Diao, Siting Li et al.ICML 2026
- Cross-Slice Knowledge Transfer via Masked Multi-Modal Heterogeneous Graph Contrastive Learning for Spatial Gene Expression InferenceZhiceng Shi, Changmiao Wang, Jun Wan, Wenwen MinCVPR 2026 · 1 citation
