Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Guy Gadboit Profile picture
Guy Gadboit
@gadboit
Pronouns: who/me?
2 subscribers
Nov 22, 2025 • 39 tweets • 9 min read
Thread on the SARS-CoV-2 reads found in the Antarctic soil samples, and what they might be able to tell us about SARS-CoV-2 origins. Very quick background: some researchers sent some of their Antarctic soil samples to Sangon Biotech, a lab also used by the Wuhan... Image Institute of Virology, in December 2019 for sequencing. They were interested in whatever they might find in Antarctic soil (nothing to do with SC2). They got the results back in "early Jan 2020". The raw reads for that sequencing run are available, and are contaminated with...
Sep 22, 2025 • 9 tweets • 3 min read
"Even only considering the two years of the pandemic [the risk of death from a super-volcanic eruption] is likely larger than the Covid risk to the woman pictured"

Brilliant new paper about the pandemic response. Thread... Image There are excellent discussions of excess death (which they estimate at around 95,000 in UK), pandemic modelling, and the actual Covid and Long Covid risk, but this graph particularly stood out for me because I kept banging on about this at the time... Image
Apr 18, 2025 • 14 tweets • 4 min read
Thread on this comprehensive new report on the subject of Pangolin viruses, the WIV, and SARS-CoV-2 origins. Although I am not convinced by the claims and conclusions myself it doesn't change the fact that this is a very useful and excellently researched reference on the... subject. Really there are two main reasons why you might think Pangolin coronaviruses (PCoVs) were involved in the origin of SC2: intriguing genetic similarities suggesting some kind of (natural or artificial) horizontal gene transfer at some point; and the fact that we know...
Mar 31, 2025 • 17 tweets • 5 min read
Here are the initial results from some further investigation I've been doing into TT and CC sequences in GISAID. Brief background: the first letter is the nucleotide found at 8782, and the second the one at 28144. Early in the pandemic there were two large lineages... Image "Lineage A" (which was TC) and "Lineage B" (which was CT, and went on to dominate everywhere). TC is closer to the bat and pangolin outgroup, and so it probably evolved into Lin B via either TT or CC. But other alternative early histories are possible. There is also a problem...
Mar 1, 2025 • 17 tweets • 4 min read
💣Brilliant new preprint by @nizzaneela explaining the major flaw in the Pekar 2022 paper. It's quite terse and technical but I will try to explain it in plain English (as best as I have understood it!) The basic idea of the original paper was to assess whether observations... about about how early sequences were related to each other was more or less likely to have happened if there was a single introduction of one version of the virus into the human population, or whether two very slightly different variants were introduced at the same time...
Feb 28, 2025 • 10 tweets • 3 min read
Update to my earlier discussion of Pekar 2025, looking in more detail at the 2020 SARS2 sequences in GISAID. To recap, the claim in the paper is that "TT" genomes (those that have C8782T but not T28144C) are more likely to be accidental later reversions than relics of an... Image earlier lineage intermediate which they reckon didn't exist. The paper also claims that if they were from an earlier lineage they should have accumulated more mutations. If those claims are true then we should see similar frequencies of *other* silent C->T mutations, not...
Feb 6, 2025 • 5 tweets • 2 min read
This is interesting. S50L, which is a reversion from SARS-CoV-2's unique 'S' to the 'L' that every single one of the bat and pangolin viruses have (represented by 5 different codons, an indicator of how much they "like" the L) seems to have started doing quite well soon after... Image I posted this thread last year in which I wondered exactly this: rather than Spike:50S being an adaptation to an intermediate host, or to a respiratory environment, maybe it was just a relatively mildly deleterious mutation that would eventually find...

Nov 16, 2024 • 13 tweets • 3 min read
Is SARS-CoV-2 also a bacteriophage? In other words, besides infecting human and mammal cells can it also infect bacteria? That's the hypothesis being investigated in this paper. Using electron microscopes, nitrogen isotopes and a plaque assay... Image
Image With the microscopes they saw what looked like viral particles sticking to bacteria, and also inside them. This was confirmed when they "labelled" the viral particles with an antibody that tags the nucleocapsid protein. But just because viral particles are getting inside...
Jun 5, 2024 • 20 tweets • 5 min read
💣I think I might have found some genetic evidence that tells us about the origin of SARS-CoV-2. Under the best hypothesis I can think of, it points to a natural origin. Let me explain. While investigating the Florianópolis wastewater reads I made the surprising discovery... Image that in three places the reads matched Pangolin viruses. Not only that, but in those locations where they matched the Pangolin viruses, Wuhan-Hu-1 had the same thing as the much more closely-related bat viruses RaTG13 and the BANALs. This "Pangolin Paradox" is hard to explain...
May 29, 2024 • 16 tweets • 4 min read
Is the lineage in the Florianópolis wastewater sample early or later contamination? I have put some of the key alleles apparently found in the Florianópolis sample, together with others from Brazil, and other early samples into a table (link at the end)... Image Some thoughts. First it looks highly likely that the Santa Catarina and Bahia human samples are related to what was found in the WW. They share many alleles, including the rare A5706G. Note however that this lineage continued in Brazil until 2020-09-03...
May 22, 2024 • 16 tweets • 4 min read
SARS-CoV-2 was somewhat prevalent in Florianópolis, Brazil already by November 27 2019, two months before the first confirmed case in the Americas, and three before the first official case in Brazil. Not just PCR, but some direct metagenomics, covering at least 7% of the... Image genome. Could it be a false positive caused by contamination? If not can it tell us anything about the early evolution of SARS-CoV-2? I don't think this is a false positive. There were positive and negative controls, and the results were confirmed by two labs and at least... Image
Mar 28, 2024 • 16 tweets • 4 min read
OK I think I've figured out what @SolidEvidence has found here: basically nothing. Diddly squat. Yes afraid so. Here's why. Let's look at the first "Cryptic Reversion" here, K1795Q. What this table is showing is that where SARS-CoV-2's bat relatives have Q, it has a K... Image But 56.5% of the time, in "cryptic lineages" (ones found in wastewater, likely persistent GI infections) that K tends to mutate back to a Q. When we look at an alignment of SARS-CoV-2 and a large list of its bat and pangolin relatives, that K (on the first row) is striking... Image
Dec 13, 2023 • 13 tweets • 3 min read
New preprint about "Original Antigenic Sin" which is interesting because they tested the thing we very rarely see confirmed: what does the immune response of somebody who started with Omicron, but hadn't had any original strain vaccines look like?... Image What's really happening with "OAS" and "Imprinting" is just your immune system finding a compromise. If you are exposed to two very similar things close together your body tries to find antibodies that neutralize *both* of them quite well...
Dec 9, 2023 • 11 tweets • 3 min read
Follow-up to this thread since people asked me to quantify my claim that you would have to be "pretty unlucky" for any of these rogue proteins to cause autoimmunity, which is something we can do a basic estimate of for ourselves... The authors of the paper created a "peptide pool" of suspected strange proteins that the Pfizer vaccine might make, and found about 25% of vaccinees for positive for something out of this pool...
Dec 3, 2023 • 14 tweets • 4 min read
Gigavaxxing doesn't work. This is a study of people who had had around 5, 6 or 7 vaccine doses altogether, the first 3 or 4 of which were Wuhan-Hu-1, and then either a bivalent, or a bivalent then the monovalent XBB.1.5 booster. They measured neutralization... Image which correlates fairly well with vaccine efficacy against symptomatic infection. None of them could neutralize any of the omicron variants worth a damn. The variant boosters mainly boosted their ability to neutralize the long-extinct strain they are never going to...
Sep 14, 2023 • 13 tweets • 4 min read
Is SARS-CoV-2 a reverse genetics system? It sure looks like one. But that doesn't mean it is. The question we are concerned with here is what are the chances that something that looked as much like one just evolved by chance? I have done a very similar analysis to the one... in this preprint, but I got very different results. I find about 45% of simulated RatG13 mutants satisfy the criteria, and about 5% of BANAL-52, in contrast to the reported numbers of 1.2% and 0.1%. My aim was not...
doi.org/10.1101/2022.1…
Image
Aug 30, 2023 • 21 tweets • 5 min read
Update to this thread, because I have made a lot of improvements to the crude grouping analysis I mentioned there, and it now shows some results. I think we can say with moderate confidence that there are more than one different viral genomes in these reads, whether from... different viruses or different variants of the same one. Rather than grouping reads according to some minimum alignment between them I am now grouping them according to a minimum *consecutive* alignment, in other words, a minimum overlap. And rather than expecting two groups...
Aug 26, 2023 • 25 tweets • 6 min read
Does the furin cleavage site in SARS-CoV-2 originate from these Pangolin samples? The samples are from 2019 but were sequenced in Guangzhou after the pandemic started, and a limited set of raw reads uploaded to ncbi here...

ncbi.nlm.nih.gov/sra/SRR1109326…
Image Those reads contain one that is a 49nts long, is a perfect match with SARS-CoV-2, and contains the furin cleavage site (FCS). It's effectively impossible that this is a coincidence but is it contamination that was introduced in early 2020 when the samples were sequenced, by...
Aug 17, 2023 • 9 tweets • 2 min read
I hadn't seen this before. Small sample, and they are cancer patients, but no reason to think you wouldn't find similar results in other people. After the third dose of Moderna (I don't know how far apart they were), 30% of the patients showed a reduced T-cell response and... T-cell markers indicative of exhaustion. T-cell exhaustion sounds bad and it kind of is in this case. It is basically a kind of tolerance mechanism for T-cells to give up attacking something because they've seen too much of it. This is good if they're attacking the wrong thing...
Jun 9, 2023 • 21 tweets • 5 min read
Going back to this discussion about Covid vaccines and HPV vaccines there is another point I didn't make at the time, because I didn't realize how important it was. The case for HPV vaccines *is* much stronger than that for Covid vaccines but you have to be... clear about why. Dr Walsh made two statements in favour of HPV vaccines which were both false but which, to anyone reflecting on them for more than a few minutes, would argue *against* the vaccines. The first was that we can rest assured that we will have long-term immunity...
Jun 7, 2023 • 7 tweets • 2 min read
Interesting discussion about the HPV vaccine. This is not a vax vs antivax debate, but a more productive discussion between the "it's a no-brainer, home-run" school of thought, represented by Dr Kristen Walsh...

open.spotify.com/episode/74xE0n… and the "centrist" view that more data and openness is needed about the risks and benefits and how those vary with the individual, represented by @KrugAlli. The starting point for the discussion was that after the breakdown of public trust caused...