Saved in:
Bibliographic Details
Main Authors: Jixin Liu, Shuli Sun, Zhengliang Lv, Xinyu Liu, Yihua Wang, Bingqiang Liu
Format: Artículo Open Access
Published: Wiley 2026
Subjects:
Online Access:https://advanced.onlinelibrary.wiley.com/doi/10.1002/advs.75607
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1867018343821606912
author Jixin Liu
Shuli Sun
Zhengliang Lv
Xinyu Liu
Yihua Wang
Bingqiang Liu
author_facet Jixin Liu
Shuli Sun
Zhengliang Lv
Xinyu Liu
Yihua Wang
Bingqiang Liu
Jixin Liu
Shuli Sun
Zhengliang Lv
Xinyu Liu
Yihua Wang
Bingqiang Liu
collection Wiley Open Access
contents STAID: A Self‐Refining Deep Learning Framework for Spatial Cell‐Type Deconvolution with Biologically Informed Modeling Jixin Liu Shuli Sun Zhengliang Lv Xinyu Liu Yihua Wang Bingqiang Liu Advanced Science ABSTRACT Spatial transcriptomics provides high‐throughput measurement of gene expression while retaining spatial context; however, inferring accurate cell‐type compositions within individual spots remains a major challenge. Here, we present STAID, a unified framework that effectively integrates pseudo‐spot generation with deep learning training through iterative pseudo‐spot refinement and leverages graph signal processing to capture higher‐order gene‐wise relationships. By creating a self‐reinforcing cycle, STAID enables accurate spot‐level deconvolution of cell‐type compositions for spatial transcriptomics data. Comprehensive benchmarking demonstrates that STAID outperforms existing methods, accurately reconstructs cell‐type spatial distributions, and effectively resolves the cellular colocalization. In clinical breast cancer sections, STAID precisely infers tumor epithelial distributions and reveals their spatial associations with immune cells. In human embryonic limb datasets, STAID captures the ordered spatial distributions of key progenitor populations, reflecting hierarchical tissue organization and demonstrating that incorporating cell‐type composition information can enhance tissue segmentation. STAID also resolves the spatial cellular organization in Crohn's disease and reveals the characteristics of TLS‐like immune niches. Collectively, by delivering high‐resolution cell‐type distributions, STAID provides deeper insights into tissue organization and cellular heterogeneity. 10.1002/advs.75607 http://creativecommons.org/licenses/by/4.0/
doi_str_mv 10.1002/advs.75607
format Artículo Open Access
id wiley_oa_10_1002_advs_75607
institution Wiley Open Access
license_str_mv http://creativecommons.org/licenses/by/4.0/
publishDate 2026
publisher Wiley
record_format wiley_oa
spellingShingle STAID: A Self‐Refining Deep Learning Framework for Spatial Cell‐Type Deconvolution with Biologically Informed Modeling
Jixin Liu
Shuli Sun
Zhengliang Lv
Xinyu Liu
Yihua Wang
Bingqiang Liu
Advanced Science
STAID: A Self‐Refining Deep Learning Framework for Spatial Cell‐Type Deconvolution with Biologically Informed Modeling Jixin Liu Shuli Sun Zhengliang Lv Xinyu Liu Yihua Wang Bingqiang Liu Advanced Science ABSTRACT Spatial transcriptomics provides high‐throughput measurement of gene expression while retaining spatial context; however, inferring accurate cell‐type compositions within individual spots remains a major challenge. Here, we present STAID, a unified framework that effectively integrates pseudo‐spot generation with deep learning training through iterative pseudo‐spot refinement and leverages graph signal processing to capture higher‐order gene‐wise relationships. By creating a self‐reinforcing cycle, STAID enables accurate spot‐level deconvolution of cell‐type compositions for spatial transcriptomics data. Comprehensive benchmarking demonstrates that STAID outperforms existing methods, accurately reconstructs cell‐type spatial distributions, and effectively resolves the cellular colocalization. In clinical breast cancer sections, STAID precisely infers tumor epithelial distributions and reveals their spatial associations with immune cells. In human embryonic limb datasets, STAID captures the ordered spatial distributions of key progenitor populations, reflecting hierarchical tissue organization and demonstrating that incorporating cell‐type composition information can enhance tissue segmentation. STAID also resolves the spatial cellular organization in Crohn's disease and reveals the characteristics of TLS‐like immune niches. Collectively, by delivering high‐resolution cell‐type distributions, STAID provides deeper insights into tissue organization and cellular heterogeneity. 10.1002/advs.75607 http://creativecommons.org/licenses/by/4.0/
title STAID: A Self‐Refining Deep Learning Framework for Spatial Cell‐Type Deconvolution with Biologically Informed Modeling
topic Advanced Science
url https://advanced.onlinelibrary.wiley.com/doi/10.1002/advs.75607