CONTAMINATION OR VACCINE RESEARCH? RNA Sequencing data of early COVID-19 patient samples show abnormal presence of vectorized H7N9 hemagglutinin segment
A re-analysis of the meta-transcriptome data (SRR11092059-63) generated at the Wuhan Institute of Virology (WIV) from bronchial-alveolar lavage fluid (BALF) specimens of five early SARS-CoV-2 patients (WIV02,04,05,06,07-2)2 was done. The data of these five patients had been obtained by the WIV on two different NGS machines: MiSeq and MGISEQ-2000RS (HiSeq 3000 equivalent). The MGISEQ-2000RS gave 10X more data than the MiSeq machine (5.6-12 Gb) and therefore, it was possible to see more detail. Surprisingly, all the five samples analysed by MGISEQ-2000RS machine showed the presence of a sequence H7N9 ‘Hemagglutinin A (HA) segment 4’ gene in a relatively high proportion, and in one case six- times the abundance of the SARS-CoV-2 sequences. The presence of non-SARS-CoV-2, including these influenza A genes, has been reported earlier and this data was also used in our current study for comparison and analysis The surprising finding was the HA segment 4 gene cloned in an expression vector, pVAX1, confirming previously identified vector sequences.3,4 A WIV publication documented that DNA vaccines containing H7N9 HA genes were being developed and tested in mice in WIV at the same time as the outbreak (2019-2020). In addition, all five samples showed a relatively high proportion of Spodoptera frugiperda rhabdovirus (13-83% of SARS-CoV-2 reads). Additionally, the samples also showed the presence of other low-abundance, high homology (LAHH) sequences, mostly of viral origin and not expected to be associated with human BALF specimens. These LAHH sequences could be contaminants, and we identified these viruses as part of previously published research at the WIV, providing a genomic record of prior work. The ability to identify previously performed research in the meta-transcriptome raw data reads from a laboratory provides a new forensic tool. The presence of cloned H7N9 HA gene segment in the transcriptome data of the early five patients processed in WIV, should be treated as an important forensic clue and warrants a full investigation. An understanding of the relationship, if any, between the presence of the vectorized H7N9 and SARS-CoV-2 in the early COVID-19 patients’ samples should be a focus of further inquiry.