Deep-learning enabled phenomics applied to COVID-19 drug discovery

We applied deep-learning-driven analysis of cellular morphology to develop a scalable “phenomics” platform and here we demonstrate its ability to rapidly identify potential therapeutic stop-gaps for the COVID-19 pandemic. High-throughput screening on this platform demonstrates rapid identification and triage of hits for COVID-19. We deploy the platform to develop phenotypic models of active SARS-CoV-2 infection and of COVID-19-associated cytokine storm, surfacing compounds with demonstrated clinical benefit, such as Remdesivir, and deprioritizing compounds that had no benefit such as hydroxychloroquine. The associated library of images, deep learning features, and compound screening data from COVID-19 screens are available at rxrx.ai and serve as a resource for immune biology and cellular-model drug discovery with potential impact on the COVID-19 pandemic