Build structural models for several variants of SARS-CoV-2 RBD

Three-dimensional coordinates for several variants of the SARS-CoV-2 receptor binding domain at the PDB format. These are build models obtained from the comparative modeling of the protein three-dimensional wildtype (wt) structure built up at the SWISS-MODEL workspace (YP_009724390.1) based on the NCBI reference sequence NC_045512. The coordinates for these variants were modelled by the simple replacement of amino acids from the SARS-CoV-2 S RBD wt structure (from the structure given by YP_009724390.1) followed by an energy minimization using “UCSF Chimera 1.14”. Rotamers using the Dunbrack 2010 library with the highest probabilities were selected for each case. The minimization was performed with default parameters also considering the H-bonds. Replaced amino acids are a) N501Y, K417N and E484K, for the South African (SA) variant, b) K417T, E484K, N501Y, for the Brazilian (BR) P.1, c) Y453F for the “mink” strain, d) N501Y, for the UK B.1.1.7 variant, e) E484K and N501Y, for the New York (NY) B.1.526 variant, f) L452R, for the Californian (CA) B.1.1.376 strain and g) E484Q and L452R, for the in Indian variant (B.1.617). Their RMSDs (Root-mean-square deviations) in comparison with the wildtype structure are BR: 0.096Å, CA: 0.113Å, E484K: 0.102Å, India: 0.157Å, mink: 0.093Å, NY: 0.094Å, SA: 0.043Å and UK: 0.094Å. These values were also calculated with “UCSF Chimera 1.14”.