, , , , ,
Highlights
- Workflow to build spatial objects linking stitched images with transcriptomics data
- Steps to apply the transformer and GraphVAE for spatial and contextual tissue modeling
-
Guide to merge gene expression and histology data into one spatial framework.
- Instructions for clustering and deconvolution to map tumor niches at the single-cell level
Summary
We present a transformer and a graph variational autoencoder to identify microenvironments (TG-ME). This computational framework integrates transformer and graph variational autoencoders to dissect spatial niches using spatial transcriptomics and morphological images. This protocol outlines data normalization, spatial transcriptomics integration, morphological feature extraction, and niche profiling. Using deep learning, TG-ME enables robust niche clustering applicable to healthy, tumor, and infected tissues.