Fast assessment of human receptor-binding capability of 2019 novel coronavirus (2019-nCoV)

Qiang Huang; Andreas Herrmann

With the complex model in Fig. 2C , more than 3,000 independent docking runs were performed to identify thermodynamically probable, low-energy binding conformations. The results are illustrated in When comparing with the SARS-CoV results in Fig. 1A , we found that there is no binding . CC-BY-NC-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.02.01.930537 doi: bioRxiv preprint 8 conformation with a binding energy score less than that of the lowest-energy conformation of SARS-CoV derived from experimentally determined structure (i.e., -37.4 kcal·mol -1 in Fig. 1A) . To be sure, we also carried out high-resolution all-atom refinements to further optimize the lowest-energy binding conformation by 200 independent runs (see Methods). As indicated the scatter plot in Fig. 4A , the lowest-energy value of the 200 refined models remains about -28.4 kcal·mol -1 (Fig. 4B) . Considering that the protein-protein docking has already sampled a large number of thermodynamically probable, low-energy binding conformations, these results indicate that the existence of an ACE2-binding conformation for 2019-nCoV S-RBD with an energy value < -37.4 kcal·mol -1 is very limited. So, the binding strength of 2019-nCoV S-RBD to ACE2 is unlikely greater than that of SARS-CoV S-RBD. versus RMSDs of interface residues of the refined models, where the RMSD reference is the lowest-energy conformation in Fig. 3A. (B) The lowest-energy model of S-RBD binding to ACE2 with an energy score of -28.4 kcal·mol -1 .