The novel coronavirus SARS-CoV-2 is known to cause COVID-19 and binds to human ACE2 receptors via its spike protein (S protein) in the receptor-binding domain (RBD). The large number of single nucleotide polymorphisms (SNPs) that can occur in the RBD can cause greater transmissibility of the virus; the unpredictable nature of SNPs makes it necessary for mutations to be characterized quickly. Measuring binding free energies (BFEs) of new virial mutations provides insights on characteristics of these new strains, and importantly, how they impact host-virus interactions. While experimental investigations can be useful to determine the BFEs of SNPs, these methods are often costly, laborious, and difficult to scale up. Thus, it is beneficial to have computational methods that simulate SNPs and mutations in multiple residues to determine how possible structural changes impact receptor-protein binding.<br/><br/>Current computational models exist based on the end-point free energy approach among which molecular mechanics generalized Born surface area, MMGB/SA, is known to be an efficient method. Through MMGB/SA, it is possible to decompose the total free energy of a wide range of biomolecular structures into sub-components and measure their contributions separately. However, a lack of accuracy hinders vast investigation of these models in real-world applications. Herein, we aim to build an accurate MMGB/SA model and compare computed BFEs with those calculated through deep mutational scanning techniques. To this end, we create a fully automated and open-source pipeline that reports mutation BFEs. Multidimensional optimization and model customization have been performed to improve the accuracy while maintaining the speed of the pipeline. This optimized model has been used for studying the wild type virus and computations were compared against experimental values. The BFE for the wildtype was found to be -9.4 ± 1.5 kcal/mol, which is in close agreement with the experimental -10.6 kcal/mol. The proposed pipeline will have great accessibility for various groups that want to study SARS-CoV-2 mutations in the future.

An automated pipeline for studying binding affinity of SARS-CoV-2 variants and the human ACE2 receptor

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.