Slack / Private Message Drop — page 177

[Andrew Rambaut] Just had the following email. It seems reasonable to get to the bottom of what is up with RaTG13 (I imagine it was sloppy work under extreme pressure). Obviously RaTG13 isn't the whole story and the nutters' obsession with it is misplaced (the existence of the pangolins and RmYN02 tell you almost as much). ```Dear Andrew, I was recently contacted by a national fact-checking body, run by Norwegian media outlets. The topic was the conspiracy surrounding an alleged release of sars-cov-2 from a Chinese lab. They pointed me to this so-called Yan-report, detailing how sars-cov-2 could possibly have been generated by functional genetics. The "Yan-report" seems obsessed with RaTG13, and cite a number of preprints (most crap, but some less so) questioning the veracity of the sequence. As RaTG13 is kind of central in arguments re the origin of sars-cov-2, I wanted to have a quick look at the sequence data. I am looking at this now together with my colleague Ola Brynildsrud, and so far, we have not been able to re-create the RaTG13 genome in GenBank from the published metagenome fastqs and sanger amplicons published with the RaTG13 genome: paper: https://www.nature.com/articles/s41586-020-2012-7 sequence data: https://www.ncbi.nlm.nih.gov/Traces/study/?acc=SRP249482&o=acc_s%3Aa I wanted to check whether any of you proper Covid people have looked into this dataset, or simply used the published GenBank sequence. The best would probably be to post the question on virological, but I dont have a user there. It might well be that the authors have been sloppy in uploading all their raw sequence data, but as said, the published raw data does not seem to be enough to generate the RaTG13 assembly, not even when we have said assembly to align reads against. But we're still working on this. Kind regards, Vegard Eldholm Norwegian Institute of Public Health``` I don't know how much further you want to get into this @Kristian Andersen but I think he is going to email you. [2020-09-17 06:44:43] [Andrew Rambaut] (I just don't know enough about sequencing bioinformatics and genome assembly to look at this). [2020-09-17 06:57:03] [Kristian Andersen] I might take a look at the sequence data - overall, I have no concerns though given that this is how sequencing of low-coverage genomes often occurs. Patchwork.... [2020-09-17 07:01:42] [Andrew Rambaut] I guess the question is has there been any 'imputation'? From the above it would seem they can't even get the consensus when they map to the actual consensus? [2020-09-17 07:03:04] [Andrew Rambaut] It would also matter which bits are low coverage. I would prefer to move on but it would be good to be armed with the knowledge - who knows where this will go next. [2020-09-17 07:10:39] [Kristian Andersen] I suspect the data is scattered across repos