10 ms·

There's nothing in that list that argues against SARS-CoV-2—and its spike protein and furin cleavage site in particular—being produced by serial passage through

There's nothing in that list that argues against SARS-CoV-2—and its spike protein and furin cleavage site in particular—being produced by serial passage through any of a variety of cell lines expressing human ACE-2. It does seem unlikely that the SARS-CoV-2 genome, or spike protein, or even just a small segment including the furin cleavage site, was synthesized. Although without knowing what protein folding modeling capabilities and synthesis capabilities the WIV had, who knows for sure?

lamontcg 5y ago
There's 1,000 nucleotide differences between the closest known progenitor and SARS-CoV-2. That is an evolutionary distance of 30-40 years in a coronavirus. You can't replicate that many years of evolution by serial passage in a lab. Serial passage isn't a magic wand. And the rest of it is that you're arguing in favor of a science fiction explanation for the capabilities of the WIV lab.
- willupowers 5y ago
  30-40 years difference from SARS-CoV. However, the proposal actually says it would use other unpublished sequences for development. These unpublished sequences may have been much closer to SARS-CoV-2. I think you should read the leaks.
  - jml7c5 5y ago
    >However, the proposal actually says it would use other unpublished sequences for development. Could you give the page # for this?
- createdapril24 5y ago
  This is a question I've encountered a few times. How quickly, versus natural evolution, can serial passage evolve a virus? Using this figure, we could have some parameters around how much time such a project must have taken and its latest start date (assuming its evolved from known or closely-related-to-known viruses). Using this figure, and a theoretical timetable, we can also specify how close a cousin virus we would need to have for such a process to develop SAR-CoV-2. This could give us a more specific understanding of the feasibility of such research. For example, RaTG13 is closest known virus was discovered in 2013. Does the rate of serial passage enable evolution from RaTG13 over ~7 years? If not, this provides a factual argument allowing us to determine even if gain-of-function research was imposed on RaTG13 the moment of its discovery, it could not have been developed into SARS-CoV-2. (I don't know how to evaluate the actual accuracy of this, but provided as a example of how knowing the serial passage rate would be helpful).
  - lamontcg 5y ago
    Serial passage isn't magic, it is just what nature does. In the lab you'll get a handful of mutations not thousands. After serial passage through some large fraction of a billion humans, with large evolutionary pressures due to the recent species jump, the delta variant is still well over 99% homologous to the reference Wu-1 strain. A 96.1% different would require serial passage through billions of organisms, but they measure this difference in terms of years of evolution in nature which is on the order of 30-40 years.
    - angelzen 5y ago
      Then 'it was not serial passage' theory is unfalsifiable (sorry for the double negation): * If a wild strain is too far away from covid, then it would have taken too long to do the passage. * If a wild strain is almost identical to covid, then obviously covid derived from that without seeing the interior of a lab. Under what circumstances would you consider that serial passage work done in a lab might have had something to do with covid?
    - passivate 5y ago
      Well said, also you have little control over which colonies will take over.
    - tomp 5y ago
      > After serial passage through some large fraction of a billion humans Wouldn’t this count as parallel passage though? Sure there’s more variability and evolutionary pressure on the mutations, but the speed of evolution (the number nucleotide mutations) is the same over the same period of time regardless of how many billions of people it infects in parallel.
    - inciampati 5y ago
      In the lab we can vastly exceed natural mutation rates, rendering this appeal to nature baseless. The rate of evolution is a function of the mutation rate and the effective population size. There is no reason to believe that a lab setting, with highly parallel evolution on a very large, diverse population, will be as slow as passage through hosts in nature. The application of mutagens, and also the lack of selection by a host immune system can support much higher rates of change. And in coronaviruses, recombination is also very frequent, and this could easily give rise to multiple % levels of divergence in a single step. Finally, it's trivial to synthesize a genome of this size, and also to synthesize pools of related viruses based on common backgrounds.

lamontcg 5y ago

There's 1,000 nucleotide differences between the closest known progenitor and SARS-CoV-2. That is an evolutionary distance of 30-40 years in a coronavirus. You can't replicate that many years of evolution by serial passage in a lab. Serial passage isn't a magic wand. And the rest of it is that you're arguing in favor of a science fiction explanation for the capabilities of the WIV lab.

willupowers 5y ago
30-40 years difference from SARS-CoV. However, the proposal actually says it would use other unpublished sequences for development. These unpublished sequences may have been much closer to SARS-CoV-2. I think you should read the leaks.
- jml7c5 5y ago
  >However, the proposal actually says it would use other unpublished sequences for development. Could you give the page # for this?
createdapril24 5y ago
This is a question I've encountered a few times. How quickly, versus natural evolution, can serial passage evolve a virus? Using this figure, we could have some parameters around how much time such a project must have taken and its latest start date (assuming its evolved from known or closely-related-to-known viruses). Using this figure, and a theoretical timetable, we can also specify how close a cousin virus we would need to have for such a process to develop SAR-CoV-2. This could give us a more specific understanding of the feasibility of such research. For example, RaTG13 is closest known virus was discovered in 2013. Does the rate of serial passage enable evolution from RaTG13 over ~7 years? If not, this provides a factual argument allowing us to determine even if gain-of-function research was imposed on RaTG13 the moment of its discovery, it could not have been developed into SARS-CoV-2. (I don't know how to evaluate the actual accuracy of this, but provided as a example of how knowing the serial passage rate would be helpful).
- lamontcg 5y ago
  Serial passage isn't magic, it is just what nature does. In the lab you'll get a handful of mutations not thousands. After serial passage through some large fraction of a billion humans, with large evolutionary pressures due to the recent species jump, the delta variant is still well over 99% homologous to the reference Wu-1 strain. A 96.1% different would require serial passage through billions of organisms, but they measure this difference in terms of years of evolution in nature which is on the order of 30-40 years.
  - angelzen 5y ago
    Then 'it was not serial passage' theory is unfalsifiable (sorry for the double negation): * If a wild strain is too far away from covid, then it would have taken too long to do the passage. * If a wild strain is almost identical to covid, then obviously covid derived from that without seeing the interior of a lab. Under what circumstances would you consider that serial passage work done in a lab might have had something to do with covid?
  - passivate 5y ago
    Well said, also you have little control over which colonies will take over.
  - tomp 5y ago
    > After serial passage through some large fraction of a billion humans Wouldn’t this count as parallel passage though? Sure there’s more variability and evolutionary pressure on the mutations, but the speed of evolution (the number nucleotide mutations) is the same over the same period of time regardless of how many billions of people it infects in parallel.
  - inciampati 5y ago
    In the lab we can vastly exceed natural mutation rates, rendering this appeal to nature baseless. The rate of evolution is a function of the mutation rate and the effective population size. There is no reason to believe that a lab setting, with highly parallel evolution on a very large, diverse population, will be as slow as passage through hosts in nature. The application of mutagens, and also the lack of selection by a host immune system can support much higher rates of change. And in coronaviruses, recombination is also very frequent, and this could easily give rise to multiple % levels of divergence in a single step. Finally, it's trivial to synthesize a genome of this size, and also to synthesize pools of related viruses based on common backgrounds.