评价一下：The rapid development of spatial transcriptomics allows the measurement of RNA abundance at a high spatial resolution making it possible to simultaneously profile gene expression spatial location

The development of spatial transcriptomics has greatly advanced the understanding of gene expression at a high spatial resolution. This paper introduces Hist2ST, a deep learning-based model that predicts gene expression from histology images. The model uses a combination of convolutional, Transformer, and graph neural network modules to extract 2D vision features and capture spatial relations between spots. The learned features are then used to predict gene expression. The model outperforms existing methods in terms of both gene expression prediction and spatial region identification, and further pathway analysis suggests that it preserves biological information. Overall, Hist2ST is a promising tool for generating spatial transcriptomics data from histology images and understanding molecular signatures of tissues