https://www.selleckchem.com/products/bgj398-nvp-bgj398.html Coronaviruses make use of a large envelope protein called spike (S) to engage host cell receptors and catalyze membrane fusion. Because of the vital role that these S proteins play, they represent a vulnerable target for the development of therapeutics. Here, we describe the isolation of single-domain antibodies (VHHs) from a llama immunized with prefusion-stabilized coronavirus spikes. These VHHs neutralize MERS-CoV or SARS-CoV-1 S pseudotyped viruses, respectively. Crystal structures of these VHHs bound to their respective viral targets reveal two distinct epitopes, but both VHHs interfere with receptor binding. We also show cross-reactivity between the SARS-CoV-1 S-directed VHH and SARS-CoV-2 S and demonstrate that this cross-reactive VHH neutralizes SARS-CoV-2 S pseudotyped viruses as a bivalent human IgG Fc-fusion. These data provide a molecular basis for the neutralization of pathogenic betacoronaviruses by VHHs and suggest that these molecules may serve as useful therapeutics during coronavirus outbreaks. Accurate modeling of the effects of mutations on protein stability is central to understanding and controlling proteins in myriad natural and applied contexts. Here, we reveal through rigorous quantitative analysis that stability prediction tools often favor mutations that increase stability at the expense of solubility. Moreover, while these tools may accurately identify strongly destabilizing mutations, the experimental effect of mutations predicted to stabilize is actually near neutral on average. The commonly used "classification accuracy" metric obscures this reality; accordingly, we recommend performance measures, such as the Matthews correlation coefficient (MCC). We demonstrate that an absurdly simple machine-learning algorithm-a neural network of just two neurons-unexpectedly achieves high classification accuracy, but its inadequacies are revealed by a low MCC. Despite the above limitations