ST2HE: A Cross-Platform Framework for Virtual Histology and Annotation of High-Resolution Spatial Transcriptomics...

Area of Science:

• Computational pathology
• Genomics
• Bioinformatics

Background:

• High-resolution spatial transcriptomics (HR-ST) provides deep insights into tissue architecture.
• Standardized histological annotation frameworks for HR-ST data are currently lacking.

Purpose of the Study:

• To introduce ST2HE, a cross-platform generative framework for synthesizing virtual hematoxylin and eosin (H&E) images from HR-ST data.
• To enable histologically faithful image generation across diverse tissue types and HR-ST platforms.
• To support downstream annotations of tissue histology and phenotype classification.

Main Methods:

• ST2HE integrates nuclei morphology and spatial transcript coordinates using a one-step diffusion model.
• Conditional and tissue-independent variants were developed to support known and novel tissue contexts.
• Evaluations were performed on breast cancer, non-small cell lung cancer, and Kaposi's sarcoma datasets.

Main Results:

• ST2HE successfully generated histologically faithful virtual H&E images from HR-ST data.
• The framework preserved morphological features crucial for downstream annotations.
• Ablation studies identified key factors for enhancing image fidelity, including context window size and loss function balancing.

Conclusions:

• ST2HE effectively bridges the molecular and histological domains in spatial transcriptomics data.
• The framework offers interpretable and scalable annotation solutions for HR-ST data.
• ST2HE represents a significant advancement in computational pathology and the analysis of tissue architecture.