7 ms·

The most likely research-origin scenario is either a naturally-evolved novel virus collected and accidentally released by the WIV, or a chimera of multiple such

The most likely research-origin scenario is either a naturally-evolved novel virus collected and accidentally released by the WIV, or a chimera of multiple such viruses. No genomic evidence could distinguish that from natural spillover. For example, here's David Relman back in 2020: > This argument [that SARS-CoV-2 must be natural since it doesn't use a known backbone] fails to acknowledge the possibility that two or more as yet undisclosed ancestors (i.e., more proximal ancestors than RaTG13 and RmYN02) had already been discovered and were being studied in a laboratory—for example, one with the SARS-CoV-2 backbone and spike protein receptor-binding domain, and the other with the SARS-CoV-2 polybasic furin cleavage site. It would have been a logical next step to wonder about the properties of a recombinant virus and then create it in the laboratory. https://www.pnas.org/doi/10.1073/pnas.2021133117 https://www.pnas.org/doi/10.1073/pnas.2021133117 The WIV had the world's largest effort to collect novel sarbecoviruses from nature, often from remote caves that no other humans would routinely visit; so I don't see any reason to discount that possibility.

bayesian_horse 4y ago
You don't see the reason because you think like a computer scientist. Having two NPM packages whose versions interfere with each other is often called "dependency hell". But it is way worse for proteins in a Virus. In order to make a pandemic Virus the components must be finely tuned to each other. That is only possible through massively parallel evolution. A lab can't do that. When a lab combines Viruses, it's mostly to learn the individual contributions of the proteins. These Viruses are way less "fit" overall.
- tripletao 4y ago
  Except that we're currently discussing an article about researchers who achieved gain-of-function in a laboratory recombinant of a pandemic virus? Their chimera has the immune escape of omicron, with most of the mortality of the Wuhan wild type. The decrease in mortality compared to the Wuhan wild type is both statistically insignificant and irrelevant regardless; the correct comparison is against natural omicron, since that's the only natural virus with the omicron spike's immune escape properties. A subsequent BU press release denies that the research involved gain of function, with bizarre arguments including the fact that since it's never been tested in humans (good?) we can't know it's actually more lethal in humans. This is obviously wrong, as many people have pointed out. For example, here's Marc Lipsitch at Harvard: https://twitter.com/mlipsitch/status/1582574061062098946?cxt=HHwWhMDUxaiut_YrAAAA https://twitter.com/mlipsitch/status/1582574061062098946?cxt... My original quote is from David Relman, a professor of microbiology at Stanford who I wouldn't expect to "think like a computer scientist". Laboratories have been enhancing potential human pandemic pathogens for literally decades now, as measured by reproduction rates in vitro, mortality in their lab animals, and other similar metrics, and publishing in the open literature. For example, here's Relman asking Baric a question back in 2014: https://www.youtube.com/watch?v=Aw-nR6-4kQQ&t=2466s https://www.youtube.com/watch?v=Aw-nR6-4kQQ&t=2466s If you were really unware that labs are successfully enhancing potential human pandemic pathogens, then a Google Scholar search for Baric will find many examples. Anyways, even if you believe (quite falsely) that laboratory gain-of-function on such pathogens is impossible, that doesn't exclude a research-related origin for SARS-CoV-2. That could be a naturally-evolved virus, but one that was sitting harmlessly in a remote cave until the WIV sampled and accidentally released it. The WIV sampled remote areas that no other humans routinely entered, so it's possible in that scenario that the lineage would have died out in bats (or other zoonotic hosts) if the WIV hadn't released it into humans. Nobody is asserting that genomic evidence shows SARS-CoV-2 to be natural anymore. Even the strongest proponents of zoonotic origin have moved on, to more complicated epidemiological and phylogenetic arguments like Pekar's paper.

bayesian_horse 4y ago

You don't see the reason because you think like a computer scientist. Having two NPM packages whose versions interfere with each other is often called "dependency hell". But it is way worse for proteins in a Virus. In order to make a pandemic Virus the components must be finely tuned to each other. That is only possible through massively parallel evolution. A lab can't do that. When a lab combines Viruses, it's mostly to learn the individual contributions of the proteins. These Viruses are way less "fit" overall.

tripletao 4y ago
Except that we're currently discussing an article about researchers who achieved gain-of-function in a laboratory recombinant of a pandemic virus? Their chimera has the immune escape of omicron, with most of the mortality of the Wuhan wild type. The decrease in mortality compared to the Wuhan wild type is both statistically insignificant and irrelevant regardless; the correct comparison is against natural omicron, since that's the only natural virus with the omicron spike's immune escape properties. A subsequent BU press release denies that the research involved gain of function, with bizarre arguments including the fact that since it's never been tested in humans (good?) we can't know it's actually more lethal in humans. This is obviously wrong, as many people have pointed out. For example, here's Marc Lipsitch at Harvard: https://twitter.com/mlipsitch/status/1582574061062098946?cxt=HHwWhMDUxaiut_YrAAAA https://twitter.com/mlipsitch/status/1582574061062098946?cxt... My original quote is from David Relman, a professor of microbiology at Stanford who I wouldn't expect to "think like a computer scientist". Laboratories have been enhancing potential human pandemic pathogens for literally decades now, as measured by reproduction rates in vitro, mortality in their lab animals, and other similar metrics, and publishing in the open literature. For example, here's Relman asking Baric a question back in 2014: https://www.youtube.com/watch?v=Aw-nR6-4kQQ&t=2466s https://www.youtube.com/watch?v=Aw-nR6-4kQQ&t=2466s If you were really unware that labs are successfully enhancing potential human pandemic pathogens, then a Google Scholar search for Baric will find many examples. Anyways, even if you believe (quite falsely) that laboratory gain-of-function on such pathogens is impossible, that doesn't exclude a research-related origin for SARS-CoV-2. That could be a naturally-evolved virus, but one that was sitting harmlessly in a remote cave until the WIV sampled and accidentally released it. The WIV sampled remote areas that no other humans routinely entered, so it's possible in that scenario that the lineage would have died out in bats (or other zoonotic hosts) if the WIV hadn't released it into humans. Nobody is asserting that genomic evidence shows SARS-CoV-2 to be natural anymore. Even the strongest proponents of zoonotic origin have moved on, to more complicated epidemiological and phylogenetic arguments like Pekar's paper.