Enhancing pan-cancer spatial transcriptomics at single-cell resolution with stPainter
Nature
Read full postResearchers developed stPainter, a pretrained conditional generative model that enhances sparse spatial transcriptomics data to single-cell resolution without needing matched scRNA-seq references or retraining. Applied to six cancer spatial transcriptomics datasets, stPainter improves expression profile reconstruction and enables detailed biological analyses, validated by spatial proteomics data.

