Researchers based at the City University of New York (CUNY) have designed a deep learning artificial intelligence (AI) model that can improve preclinical predictions of drug responses in humans. As outlined in the Oct. 17, 2022, online issue of Nature Machine Intelligence, the researchers believe their model – a context-aware deconfounding autoencoder (CODE-AE) – can help improve the quality of early drug response prediction and help reduce subsequent clinical trial failures.