Thor: a platform for cell-level investigation of spatial transcriptomics and histology [Bladder Cancer]

Spatial transcriptomics links gene expression with tissue morphology, however, current tools often prioritize genomic analysis, lacking integrated image interpretation. To address this, we present Thor, a comprehensive platform for cell-level analysis of spatial transcriptomics and histological images. Thor employs an anti-shrinking Markov diffusion method to infer single-cell spatial transcriptome from spot-level data, effectively combining gene expression and cell morphology. The platform includes 10 modular tools for genomic and image-based analysis, and is paired with Mjolnir, a web-based interface for interactive exploration of gigapixel images. Thor is validated on simulated data and multiple spatial platforms (ISH, MERFISH, Xenium, Stereo-seq). Thor identifies regenerative signatures in heart failure, unbiasedly screens breast cancer hallmarks, resolves fine layers in mouse olfactory bulb, and annotates fibrotic heart tissue. In high-resolution Visium HD data, it enhances spatial gene patterns aligned with histology. By bridging transcriptomic and histological analysis, Thor enables holistic tissue interpretation in spatial biology. Archival FFPE tissue blocks from muscle-invasive bladder cancer patients treated at The Houston Methodist Hospital were obtained under IRB protocol PRO00037670 (“Molecular analysis of archived cancer tissues”). FFPE tissue sections were processed using the 10x Genomics Visium HD Spatial Gene Expression platform according to the manufacturer’s protocol (Visium HD Spatial Gene Expression Reagent Kits for FFPE, CG000663). Briefly, 5 µm-thick FFPE sections were mounted onto Visium HD slides and baked at 60°C for 1 hour. Following deparaffinization and heat-induced epitope retrieval, tissue sections were hybridized with probe sets targeting the human transcriptome and subjected to reverse transcription to capture spatial barcodes and unique molecular identifiers (UMIs). .Slides were then permeabilized at 37 °C for 30 minutes to release the barcoded cDNA fragments, which were collected and amplified to generate sequencing libraries. Whole-transcriptome libraries were sequenced on an Illumina platform to a target depth of approximately 200–300 million reads per capture area. Image alignment, UMI counting and spatial mapping were performed using 10x Genomics’ Loupe Browser and Space Ranger (version 3.1.0) pipeline. Quality-control metrics were reviewed to exclude low-quality bins prior to downstream analyses. *************************************************************** Raw files for human/patient samples were not submitted to GEO due to concerns about submitting personally identifiable sequence data for open access. ***************************************************************