The total RNA of FJzz1 variants F20, F50, F100, F150 and F200 was extracted with the RNeasy Mini Kit (Qiagen, Hilden, Germany), and then reverse transcribed into cDNA with the Revert Aid First Stranded cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA), as described previously. PCR was performed with this cDNA as the template and 16 overlapping pairs of primers (Supplementary Table S1). The genomic sequences of the FJzz1 variants were determined with the next-generation sequencing technology. The aa sequences of the S proteins and the full-length genome sequences of these strains were also aligned using the Clustal W method in MegAlign (DNAStar Lasergene), and the N-linked glycosylation sites in the S proteins of these strains were predicted and analyzed with the NetNGlyc 1.0 Server, as described previously (Chen et al. 2019 (link)). A phylogenetic tree was constructed with the neighbor-joining method (NJ) using MEGA 7.0 based on S gene of FJzz1 strain and another 60 reference strains with complete S gene sequences available in GenBank (Supplementary Table S2).