DeepSpot: Leveraging Spatial Context for Enhanced Spatial Transcriptomics Prediction from H&E Images

Abstract

Spatial transcriptomics technology remains resource-intensive and unlikely to be routinely adopted for patient care soon. This hinders the development of novel precision medicine solutions and, more importantly, limits the translation of research findings to patient treatment. Here, we present DeepSpot, a deep-set neural network that leverages recent foundation models in pathology and spatial multi-level tissue context to effectively predict spatial transcriptomics from standard H&E images. DeepSpot substantially improved gene correlations across multiple datasets from patients with metastatic melanoma, kidney, lung, or colon cancers as compared to previous state-of-the-art. Using DeepSpot, we generated 3,780 TCGA virtual spatial transcriptomics samples (56 million spots) of the melanoma, renal cell cancer, lung adenocarcinoma and lung squamous cell carcinoma cohorts. We anticipate this to be a valuable resource for biological discovery and a benchmark for evaluating spatial transcriptomics models. We hope that DeepSpot and this dataset will stimulate further advancements in virtual spatial transcriptomics analysis.