Researchers at Tsinghua University and Peking Union Medical College investigate the architecture of SARS-CoV-2’s RNA in living cells using a new technology called icSHAPE. By treating cells with a chemical that recognizes single-stranded RNA, researchers could study which segments of an RNA molecule were pairing with other RNA segments in a double-strand and which mRNA segments were free-floating on their own. Using this data, they could visualize the RNA secondary structure, which is comprised of configurations like stems, loops, bulges, and pseudoknots, of multiple types of coronaviruses, including SARS-CoV-2. They then took their structural information from icSHAPE experiments and used it to train their deep-learning neural network called PrismNet, which predicted to which proteins segments of RNA would bind, and found 42 host cell proteins likely interact with the SARS-CoV-2 RNA genome. Scientists then found a subset of these proteins that could be deactivated without hurting the host cell and postulate that drugs that attack these host cell proteins that interact with SARS-CoV-2 may be potentially used against the virus.