I posted and deleted a few comments. I don't know what to make of this and it's getting controversial. If it could be helpful, I'd love to improve it. I just don't know where it fits in ongoing discussions by prominent researchers on analysis of strains and types.
Three things that jump out at me:
a) You'll get more credibility from having a thorough lit review than from listing credentials. Linking to magazine articles from journalists on genetic variations, when there's academic research to cite, is a little bewildering.
b) You don't identify something harmless by just testing asymptomatics. This is counterintuitive, but a great example of the Wason selection task:
You would probably want to survey the rate of variants in asymptomatics in a certain community, then compare that with the rate of variants in fatal cases in the same community to look for significant differences (and be fully prepared not to find any significant differences beyond chance, a real and likely possibility).
c) I'm still not sure what the author is specifically recommending we do next. Researchers are looking for functional variants, we don't have one to distribute yet, so we should do more sequencing. Ok, great. From what populations? How many geographic areas? Where do we get the samples? There are bottlenecks in sample collection, how do we overcome those? Given (b), how can we get more samples from fatal cases, to compare virulence to those in the general population? Task overloaded hospitals to send us additional samples from dying patients? Is there another way we can get this info?
I don't want to dismiss this out of hand, this might be a good area for additional focused research. I do think there are some key unanswered questions though, and currently a worrying disconnect with the current state of research on variants.
EDIT: Daniel, I know you're going through the comments here, and some of them have gotten pretty harsh, sorry for that. I genuinely hope you're on to something, and are able to continue to refine the post into a more robust and specific proposal using all this feedback, despite the tenor of some of the comments. That's really the best case for all of us, so good luck.
Just woke up (I am in Australia). I expect some harsh comments as the idea is controversial. To your specific points.
a) The post was intended for a lay audience, not other scientists directly. The reason I didn’t write a scientific paper on this is it would sink without a trace as most scientists and doctors are too conservative. It is quite a radical idea and while scientifically sound, it will take mavericks outside of science to see this done. For better or worse it will take the efforts of someone like on of the tech billionaires to push this forward.
b) If I had to choose one aspect that gets most commonly confused by people who read the idea it would be the reason for searching only mild/asymptomatic cases. This is purely an efficiency issue. In an ideal world we would sequence the strains in all cases, look to see if we can find mutants with deletions, and then see what was the clinical outcome of those infected with that strain. If we find that all cases of a particular mutant are mild/asymptomatic then we would have our candidate.
Because we live in a constrained world where it is not possible to sequence all cases, where should we first look? Since we are looking for a mutant that only causes mild/asymptomatic cases we can exclude patients with serious symptoms as a first pass. Once we find a candidate strain that has the right sort of mutation we can then sequence all cases in the local area, both serious and mild to get the full clinical picture. This will make the search much more efficient.
c) I am recommending that we put b) into action and get on with specifically looking for an attenuated (only causes mild disease) strain ASAP. If we find one then we can discuss what to do next, but the first step is to get started.
The good news is this can be done quickly and relatively cheaply provide we have the will and the support of someone with the clout to make it happen. It won’t be easy, but it can be done.
Tech billionaire for publicity, maybe. But just getting a guide ready might help more - everybody with a moderate amount of cash and access to the medical system somewhere can start doing this in his back yard.
> b) You don't identify something harmless by just testing asymptomatics.
True, but I think you are missing part of Daniel's proposal. You don't "just" check for unique variations of the virus in asymptomatic carriers, you also check the sequence for major deletions: We would ideally be looking for a virus strain with a large(ish) deletion in an essential viral gene. This sort of mutation is easy to spot in the SARS-CoV-2 genome data, and because the genetic information has been removed, it makes the virus very unlikely to be able to mutate back into a dangerous strain. Ideally, the strain identified will have infected a number of other people in the local area too so we can know it is safe.
So while you do want to verify that the sequence is not also present in "serious" cases, you are prescreening by a factor that theoretically should correlate to severity. Whether you "need" to do additional testing depends on your assessment of base risk. If you assume that almost everyone is going to be infected with a powerful strain if you do nothing, doing your testing in live cases with a strain you suspect is mild might still be justified. Ideally (as quoted from Daniel) this "testing" might be mostly satisfied by just monitoring those already infected by the original carrier(s) with the same mild strain.
8 and red. Most people don't invert the logic and realize that there is no claim that an odd number can't have a red back. I think confirmation bias accounts for missing the brown.
Three things that jump out at me:
a) You'll get more credibility from having a thorough lit review than from listing credentials. Linking to magazine articles from journalists on genetic variations, when there's academic research to cite, is a little bewildering.
b) You don't identify something harmless by just testing asymptomatics. This is counterintuitive, but a great example of the Wason selection task:
https://en.wikipedia.org/wiki/Wason_selection_task
You would probably want to survey the rate of variants in asymptomatics in a certain community, then compare that with the rate of variants in fatal cases in the same community to look for significant differences (and be fully prepared not to find any significant differences beyond chance, a real and likely possibility).
c) I'm still not sure what the author is specifically recommending we do next. Researchers are looking for functional variants, we don't have one to distribute yet, so we should do more sequencing. Ok, great. From what populations? How many geographic areas? Where do we get the samples? There are bottlenecks in sample collection, how do we overcome those? Given (b), how can we get more samples from fatal cases, to compare virulence to those in the general population? Task overloaded hospitals to send us additional samples from dying patients? Is there another way we can get this info?
I don't want to dismiss this out of hand, this might be a good area for additional focused research. I do think there are some key unanswered questions though, and currently a worrying disconnect with the current state of research on variants.
EDIT: Daniel, I know you're going through the comments here, and some of them have gotten pretty harsh, sorry for that. I genuinely hope you're on to something, and are able to continue to refine the post into a more robust and specific proposal using all this feedback, despite the tenor of some of the comments. That's really the best case for all of us, so good luck.