Protein structure and sequence re-analysis of 2019-nCoV genome does not indicate snakes as its intermediate host or the unique similarity between its spike protein insertions and HIV-1

Chengxin Zhang; Wei Zheng; Xiaoqiang Huang; Eric W. Bell; Xiaogen Zhou; Yang Zhang

The bioinformatics analysis performed in Ji et al. study has several limitations. First, there are only approximately 300 protein coding sequences (CDSs) in the NCBI GenBank for the snake species they chose from Bungarus multicinctus and Naja atra. These CDSs represent <2% of all protein coding genes in a typical snake genome; the genome of King cobra (Naja hannah), for example, encodes 18,387 proteins according to UniProt (https://www.uniprot.org/proteomes/UP000018936). The limited numbers of known CDSs in Bungarus multicinctus and Naja atra mean that the RSCU statistics may not reflect the actual RSCU distribution in the whole genome. Second, the Codon Usage Database used in the analysis of Ji et al. has not been updated since 2007; a re-analysis using more recent codon usage database such as CoCoPUTs is needed. Third, only 8 vertebrates were analyzed in their study, while there are >100,000 vertebrates with at least one CDS in the NCBI GenBank database. Finally, there is no established evidence that viruses evolve their codon usage to resemble that of their animal hosts 16 ; this calls for a careful benchmark of RSCU analysis in terms of its ability to re-discover known hosts of characterized viruses.