Spatial Transcriptomics Analysis of Gene Expression Prediction using Exemplar Guided Graph Neural Network

Spatial transcriptomics (ST) is essential for understanding diseases and developing novel treatments. It measures the gene expression of each fine-grained area (i.e., different windows) in the tissue slide with low throughput. This paper proposes an exemplar guided graph network dubbed EGGN to accurately and efficiently predict gene expression from each window of a tissue slide image. We apply exemplar learning to dynamically boost gene expression prediction from nearest/similar exemplars of a given tissue slide image window. Our framework has three main components connected in a sequence: i) an extractor to structure a feature space for exemplar retrievals; ii) a graph construction strategy to connect windows and exemplars as a graph; iii) a graph convolutional network backbone to process window and exemplar features, and a graph exemplar bridging block to adaptively revise the window features using its exemplars. Finally, we complete the gene expression prediction task with a simple attention-based prediction block. Experiments on standard benchmark datasets indicate the superiority of our approach when compared with past state-of-the-art methods. Graphical Abstract In this paper, we aim to predict gene expression of each window in a tissue slide image. Given a tissue slide image, we encode the windows to feature space, retrieve its exemplars from the reference dataset, construct a graph, and then dynamically predict gene expression of each window with our exemplar guided graph network. Highlights We propose an exemplar guided graph network to accurately predict gene expression from a slide image window. We design a graph construction strategy to connect windows and exemplars for performing exemplar learning of gene expression prediction. We propose a graph exemplar bridging block to revise the window feature by using its nearest exemplars. Experiments on two standard benchmark datasets demonstrate our superiority when compared with state-of-the-art approaches.