Alpha Spatial: Language-guided prediction of single-cell spatial gene expression from histology
- 1Tsinghua University
Abstract
Routine histology preserves tissue architecture but leaves its molecular organization largely unmeasured. Recovering this information requires learning across heterogeneous gene panels and constraining a biologically ill-posed morphology-to-expression mapping. Here we present Alpha Spatial (AS), a language-conditioned framework that predicts cell-level gene expression from histology while jointly learning from different tissues and panels. Matched biological context improved prediction over omitted context, whereas incompatible context reduced accuracy. Multi-tissue learning further improved breast-tissue prediction, and AS outperformed the evaluated methods while preserving molecular-state organization. In breast cancer, AS recovered POSTN-associated stromal programmes and HER2-associated molecular and spatial differences. Exploratory patient-level analyses identified a CXCL14-associated spatial state whose abundance was associated with non-response to trastuzumab-containing treatment and larger residual invasive tumours. These findings establish a framework for reusing heterogeneous molecular observations to investigate tissue organization and treatment-associated microenvironments in archived histology.
Related
- Octave: Scale-resolved Evaluation of Spatial Gene Expression Prediction from Histology
- MorphoProt: RNA-Conditioned Integration of Multiscale Histology for Spatial Protein Inference
- HESTIA: Scalable Multimodal Integration of Histology and High-Resolution Spatial Transcriptomics for Robust Spatial Domain Identification
- Slide-level batch structure limits histology-guided supervision of transcriptomic foundation models
- MINT infers the latent single-cell spatial transcriptome from paired histology
BibTeX
@misc{yang2026alpha,
title = {{Alpha Spatial: Language-guided prediction of single-cell spatial gene expression from histology}},
author = {X. YANG and Z. GUO and X. GUO and W. WANG and X. LIU and C. Zhang and B. Bai and Y. He and Y. LIU},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.08.756444},
url = {https://www.biorxiv.org/content/10.64898/2026.10.08.756444v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Alpha Spatial: Language-guided prediction of single-cell spatial gene expression from histology
AU - X. YANG
AU - Z. GUO
AU - X. GUO
AU - W. WANG
AU - X. LIU
AU - C. Zhang
AU - B. Bai
AU - Y. He
AU - Y. LIU
PY - 2026
DA - 2026/10/08
PB - bioRxiv
DO - 10.64898/2026.10.08.756444
UR - https://www.biorxiv.org/content/10.64898/2026.10.08.756444v1
AB - Routine histology preserves tissue architecture but leaves its molecular organization largely unmeasured. Recovering this information requires learning across heterogeneous gene panels and constraining a biologically ill-posed morphology-to-expression mapping. Here we present Alpha Spatial (AS), a language-conditioned framework that predicts cell-level gene expression from histology while jointly learning from different tissues and panels. Matched biological context improved prediction over omitted context, whereas incompatible context reduced accuracy. Multi-tissue learning further improved breast-tissue prediction, and AS outperformed the evaluated methods while preserving molecular-state organization. In breast cancer, AS recovered POSTN-associated stromal programmes and HER2-associated molecular and spatial differences. Exploratory patient-level analyses identified a CXCL14-associated spatial state whose abundance was associated with non-response to trastuzumab-containing treatment and larger residual invasive tumours. These findings establish a framework for reusing heterogeneous molecular observations to investigate tissue organization and treatment-associated microenvironments in archived histology.
ER -