Thor: a platform for cell-level investigation of spatial transcriptomics and histology

Area of Science:

• Spatial biology
• Genomics
• Bioinformatics

Background:

• Spatial transcriptomics (ST) integrates gene expression with tissue architecture.
• Current ST tools often lack integrated histological image interpretation.
• Analyzing ST and morphology at the cell level remains a challenge.

Purpose of the Study:

• To present Thor, a comprehensive platform for cell-level analysis of spatial transcriptomics and histological images.
• To bridge the gap between genomic data and image interpretation in spatial biology.
• To enable holistic tissue interpretation by combining gene expression and morphology.

Main Methods:

• Thor utilizes an anti-shrinking Markov diffusion method to infer single-cell spatial transcriptomes from spot-level data.
• The platform incorporates 10 modular tools for genomic and image-based analyses.
• Mjolnir, a web-based interface, facilitates interactive exploration of gigapixel images.

Main Results:

• Thor successfully integrates gene expression with cell morphology.
• The platform was validated on simulated data and multiple spatial platforms (ISH, MERFISH, Xenium, Stereo-seq).
• Applications include characterizing regenerative signatures in heart failure, screening breast cancer hallmarks, resolving olfactory bulb layers, and annotating fibrotic heart tissue.

Conclusions:

• Thor enables a holistic interpretation of spatial biology data by bridging transcriptomic and histological analysis.
• The platform enhances the alignment of spatial gene patterns with histology, particularly in high-resolution data.
• Thor facilitates deeper insights into tissue biology through integrated multi-modal analysis.