EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PSSD: Progressive Spatial-Semantic Decoupling for flow-based gene expression prediction from histology images

Chengyang Zhang iD, Bo Li iD, Bob Zhang, Yuansong Zeng iD, Yuhao Yi, Jiancheng Lv

Bioinformatics · 2026

DOI: 10.1093/bioinformatics/btag598 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Chengyang Zhang und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer Science, Sichuan University, Chengdu, China. Hauptthemen laut MEDLINE: Gene Expression Profiling, Computational Biology, Image Processing, Computer-Assisted, Software. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Major Scientific Instruments and Equipments Development Project of National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Predicting spatial gene expression from histology images offers a cost-effective complement to spatial transcriptomics. However, existing methods struggle to balance spatial continuity with functional heterogeneity, often producing over-smoothed predictions or neglecting spatial context. Results We present PSSD, a conditional flow matching framework with progressive spatial-semantic decoupling. PSSD models spatial and semantic information through separate but interacting pathways and employs a three-stage architecture with decoupled flows, adaptive fusion, and cross-stream coupling to generate biologically coherent, high-fidelity gene expression profiles. Across seven spatial transcriptomics datasets spanning multiple tissues and resolutions, PSSD consistently achieved the highest Pearson correlation coefficients among the compared methods while better preserving biological boundaries and spatial autocorrelation. Under the same sampling protocol, PSSD reduced inference time from 32.04 to 3.98 min per sample on DLPFC compared with the diffusion-based Stem model and achieved approximately sevenfold acceleration on the BC and cSCC datasets without compromising predictive quality. These results demonstrate that flow-based spatial-semantic decoupling provides an effective and computationally efficient bridge between histology and transcriptomics. Availability and implementation The source code and data are available at https://github.com/ChyaZhang/PSSD.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Gene Expression ProfilingComputational BiologyImage Processing, Computer-AssistedSoftwareSpatial TranscriptomicsHumansAlgorithmsSemantics

Autor:innen und Institutionen

  • College of Computer Science, Sichuan University, Chengdu, China.
  • Department of Pathology and Institute of Clinical Pathology, West China Hospital, Sichuan University, Chengdu, China.
  • Department of Computer and Information Science, University of Macau, Macau, China.
  • School of Big Data and Software Engineering, Chongqing University, Chongqing, China.

ORCID-Profile: Chengyang Zhang, Bo Li, Yuansong Zeng

Förderung

  • National Major Scientific Instruments and Equipments Development Project of National Natural Science Foundation of China

Zitieren

APA

Zhang, C., Li, B., Zhang, B., Zeng, Y., Yi, Y., & Lv, J. (2026). PSSD: Progressive Spatial-Semantic Decoupling for flow-based gene expression prediction from histology images. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag598

Vancouver

Zhang C, Li B, Zhang B, Zeng Y, Yi Y, Lv J. PSSD: Progressive Spatial-Semantic Decoupling for flow-based gene expression prediction from histology images. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag598

BibTeX
@article{Zhang2026,
  author = {Zhang, Chengyang and Li, Bo and Zhang, Bob and Zeng, Yuansong and Yi, Yuhao and Lv, Jiancheng},
  title = {PSSD: Progressive Spatial-Semantic Decoupling for flow-based gene expression prediction from histology images},
  journal = {Bioinformatics},
  year = {2026},
  doi = {10.1093/bioinformatics/btag598}
}
Kostenlos starten →