# Origins of SARS CoV-2 - split from: Covid-19 miscellany

#### Bomb#20

##### Contributor
Why should anyone care what the hack non-expert Wade says about it?
Unwillingness to read an opposing argument and an ad hominem against the person who made it: the sure sign of an open mind.
Unless you're an expert, there's nothing wrong with pointing out his non expertise, it's just good epistemology. There are expert sources available to cite on the issue.
Calling him "non-expert" was not the ad hominem. Calling him "hack" was the ad hominem.

And Bomb didn't show his math neither to say this:
Double-standard much? You didn't show any evidence that Wade is a "hack".

I didn't give any pronouncements on probabilities like you did. But here is one example of why I call Wade a hack.
Whether you had reason at the time to think he's a hack isn't the issue; I also had reason at the time to think lab-leak was probable. You didn't post your evidence and you complained I didn't post mine. (Plus I had posted mine, indirectly; it was included in the link you'd declined to read.)

#### Bomb#20

##### Contributor
Okay, just as a for-example, the two arginines in SARS2's furin cleavage site -- the novel insertion that makes the virus infectious in humans -- are coded CGG-CGG. Back of the envelope, the odds against that happening by randomly flipping RNA bases would appear to be four hundred to one. There are six different three-base RNA sequences that all code for arginine, and CGG is the rarest code for arginine in coronaviruses, used for only five percent of the arginines in SARS2. That would seem to imply likely formation by recombination rather than by point mutation. This raises the question, where were those CGGs copied from? CGG-CGG is a sequence that hasn't been found in any other beta coronavirus. But CGG is a very common code for arginine in humans. For a scientist trying to create a cleavage site in order to perform a gain-of-function experiment of the sort we know the Wuhan lab was doing, putting in two CGGs would be a natural choice.

Clearly, you're not an expert neither. You are describing some math, but you are not giving any background basis for what the inputs should be, because you don't have it. Don is right about it sounding like ID, this is no different than when Michael Behe comes up with some probability that proves evolution couldn't produce some protein. Just like Behe does, your .05 x .05 assumes simultaneous mutation. And a 1/400 number is hardly a big hurdle anyway when you consider a single covid infection produces billions of copies.

My point is that you can't even assign a probability because you have no idea of how many rolls of the dice there were.
I.e., you guys and Don didn't understand the argument. Nobody is claiming SARS1.9 couldn't have naturally evolved a furin cleavage site. This isn't a Behe argument; I'm not assuming simultaneous mutation; and the number of copies/rolls of the dice is irrelevant. This is just classic Bayesian reasoning. If one of the natural origin scenarios is what happened then the sequence would probably have been different from what it's observed to be; therefore the observation that it's CGG-CGG makes those scenarios less likely.

The trouble with a Behe argument is that he focuses on mutation and ignores natural selection. "Some protein" could have evolved gradually by getting some of the required changes, surviving due to superior fitness while its competitors died out, then getting more of the required changes, and repeat ad nauseam. For Darwinian evolution to work, the intermediate forms have to have a selective advantage over the alternative sequences being compared. In the case of the furin cleavage site, the skeptics are talking about redundancy in the genetic code. The alternatives they are comparing the observed sequence to aren't the inferior RNA sequences that code for less fit amino acid sequences, as in a Behe argument; they're comparing it specifically to the functionally equivalent alternative RNA sequences that code for the exact same amino acids. Those possible sequences aren't any less fit. So there's nothing there for natural selection to sink its teeth into: it has no mechanism to amplify the probability of the final sequence by accumulating the mutations sequentially and selecting the winners at each step. So the improbability of the observed sequence remains the same whether it formed simultaneously or serially.

The reason billions of copies make a difference to a Behe argument is because an improbable thing can of course happen if there are billions of rolls of the dice. But when that's why some improbable thing happened, inevitably an awful lot of probable things happened too. Behe can of course ask, "If the unlikely protein formed after a billion throws of the dice, then why don't we see a billion alternative forms too?", but the answer is obvious -- we don't see them because they were inferior, so they died out. But the thirty-five more likely alternatives to the observed furin cleavage site sequence aren't inferior. So if SARS2 came about because nature tried for it a billion times, we would probably have seen some of the other thirty-five possible versions too. So the improbability of the observed sequence remains the same whether it formed in one try or in a billion.

(Of course, the entire argument falls apart if somebody can exhibit some other beta coronavirus endemic in the area that already had a furin cleavage site with its arginines coded CGG-CGG. It's not an improbable sequence if it's sitting around there waiting to be copied. But as far as I've heard, no such candidate source virus has been found.)

I should add that this isn't necessarily a good reason to think lab-leak is probable; it's just an explanation for why what I wrote wasn't a Behe-style argument. Wade exhibited an improbable coincidence that natural origin involves. But during the time since I raised this issue last year, somebody has also exhibited an improbable coincidence that the lab-leak hypothesis involves. I'm happy to withhold judgment on which improbable coincidences are more improbable than which, until the recent papers are peer-reviewed.

#### Don2 (Don1 Revised)

##### Contributor
Okay, just as a for-example, the two arginines in SARS2's furin cleavage site -- the novel insertion that makes the virus infectious in humans -- are coded CGG-CGG. Back of the envelope, the odds against that happening by randomly flipping RNA bases would appear to be four hundred to one. There are six different three-base RNA sequences that all code for arginine, and CGG is the rarest code for arginine in coronaviruses, used for only five percent of the arginines in SARS2. That would seem to imply likely formation by recombination rather than by point mutation. This raises the question, where were those CGGs copied from? CGG-CGG is a sequence that hasn't been found in any other beta coronavirus. But CGG is a very common code for arginine in humans. For a scientist trying to create a cleavage site in order to perform a gain-of-function experiment of the sort we know the Wuhan lab was doing, putting in two CGGs would be a natural choice.
Clearly, you're not an expert neither. You are describing some math, but you are not giving any background basis for what the inputs should be, because you don't have it. Don is right about it sounding like ID, this is no different than when Michael Behe comes up with some probability that proves evolution couldn't produce some protein. Just like Behe does, your .05 x .05 assumes simultaneous mutation. And a 1/400 number is hardly a big hurdle anyway when you consider a single covid infection produces billions of copies.
My point is that you can't even assign a probability because you have no idea of how many rolls of the dice there were.
I.e., you guys and Don didn't understand the argument.
Nobody misunderstood your argument, they just disagreed with it as well as the facts and reasoning you used, obtained from Wade.
Bomb#20 said:
Nobody is claiming SARS1.9 couldn't have naturally evolved a furin cleavage site.
But you are claiming it is a very rare thing which is a problem for at least two reasons. (1) Your information from Wade was wrong--it's common and (2) even if it were not common, that's the point of natural evolution...in a world of things it does, the results are often an uncommon or improbable thing. Creationists often argue how unique or specific the thing is, rather than taking an outside view that there are a gazillion improbable things and one ends up as a result. It's like if you buy a lottery ticket and use the machine to select the numbers and they come back with some combination--any combination--you can't then say what were the odds of that?! I understand it's more nuanced in a sense, but in a broader view of not merely the lottery ticket, you also have a hundred people each choosing a lottery ticket and you are ignoring all of them except 1, saying, "see look?! this one here has some consecutive numbers! That was most likely intelligently designed." There are many other features that nature might evolve that would also be suspicious to conspiracy theorists. The probability is high that at least one naturally evolved feature would be cause to start screaming about intelligently designed inserts into the genome.
Bomb#20 said:
This isn't a Behe argument; I'm not assuming simultaneous mutation; and the number of copies/rolls of the dice is irrelevant.
No, the number of rolls of the dice is relevant, but perhaps not in the exact way you mean.
Bomb#20 said:
If one of the natural origin scenarios is what happened then the sequence would probably have been different from what it's observed to be;
Yes, odds are that you can find an evolved feature out of many evolved features and say, "the odds of that one thing happening that way are too rare."
Bomb#20 said:
therefore the observation that it's CGG-CGG makes those scenarios less likely.
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one. If you were actually able to look at each and every involved issue and show the probabilities, you might have some kind of point, but your assessment of one or the other using Bayesian statistics to derive a mathematical statement that P(X)>.5 is pseudoscience because you're just listing out some incomplete things, some cherry-picked stringencies by Wade, some not included, and some unknown and then saying "but I'm allowed to do it because it's intuition!" That's kind of ridiculous. It would be like me saying my probabilistic assessment based on all available evidence is intuitively that the P(X)<.25.

Since we're talking about CGG, it should be stated that CGG is common in codon usage. So IF it's naturally evolved OR recombinant from an intermediate host, then all that is required is the 100% probable past event that such intermediate host had CGG as common codon. We don't actually know the codon usage in most organisms. We don't know if mink has CGG as a common usage or not. However, we DO know it's common with hamster and that hamster could spread the virus just like mink.

There is not a lot of published data on codon usage for tons of potential intermediate hosts but CGG is expected to be common.

What I can also tell you is that the 12nt sequence is in the hamster genome. Just do a BLAST to confirm. Here is top hit:
Cricetulus griseus strain 17A/GY unplaced genomic scaffold, alternate assembly CriGri-PICRH-1.0 unplaced_scaffold_55, whole genome shotgun sequence
Query  1      CCTCGGCGGGCA  12
||||||||||||
Sbjct  85478  CCTCGGCGGGCA  85489

The point isn't to promote a hamster hypothesis, just to show it's more common than let on.

The 12nt sequence also hits to American mink. Here is top hit:
Neovison vison isolate M4711 chromosome 1, ASM_NN_V1
Query  1       CCTCGGCGGGCA  12
||||||||||||
Sbjct  920139  CCTCGGCGGGCA  920128

And here's pangolin:
Manis pentadactyla isolate MP20 unplaced genomic scaffold, YNU_ManPten_2.0 scaffold_111_MP20
Query  1       CCTCGGCGGGCA  12
||||||||||||
Sbjct  941542  CCTCGGCGGGCA  941553

Last edited:

#### Don2 (Don1 Revised)

##### Contributor
ferret top hit:
Mustela putorius furo isolate JIRA1106 unplaced genomic scaffold, ASM1176430v1.1 tig00000040_pilon_pilon_pilon
Query  1        CCTCGGCGGGCA  12
||||||||||||
Sbjct  4207733  CCTCGGCGGGCA  4207722

cat top hit:
Felis catus isolate Fca126 chromosome A1, F.catus_Fca126_mat1.0

Query  1         CCTCGGCGGGCA  12
||||||||||||
Sbjct  25383421  CCTCGGCGGGCA  25383410

white-tailed deer top hit:
Odocoileus virginianus texanus isolate animal Pink-7 unplaced genomic scaffold, Ovir.te_1.0 scaffold101
Query  1        CCTCGGCGGGCA  12
||||||||||||
Sbjct  1634209  CCTCGGCGGGCA  1634198


#### Don2 (Don1 Revised)

##### Contributor
Back to codon usage.

I was able to find mink from some digging:
CGU  4.6(    57)
CGC  8.3(   103)
CGA  5.0(    62)
CGG 11.3(   140)
AGA 10.7(   133)
AGG  9.5(   118)

CGG is common.

Here is human:
CGT 4.7
CGC 10.9
CGA 6.3
CGG 11.9
AGA 11.5
AGG 11.4

Here is Chinese hamster:
CGU  5.6
CGC  9.3
CGA  7.2
CGG 10.1
AGA 10.1
AGG 10.2

Pangolin seems not computed.

#### Angra Mainyu

##### Veteran Member
Don2 (Don1 Revised) said:
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one.
No, that is false.
First, we are talking about epistemic probability of course, and of course it's neither 0 nor 1 (or 0%/100% in your notation).

Second, with your criterion, if the universe happens to be deterministic, then the future probability would also be either 0 or 1.

Don2 (Don1 Revised) said:
If you were actually able to look at each and every involved issue and show the probabilities, you might have some kind of point, but your assessment of one or the other using Bayesian statistics to derive a mathematical statement that P(X)>.5 is pseudoscience because you're just listing out some incomplete things, some cherry-picked stringencies by Wade, some not included, and some unknown and then saying "but I'm allowed to do it because it's intuition!"
We humans make intuitive probabilistic assessments all the time. Mathematical probability is a model of that probability. It allows sometimes for greater precision, at the cost of ignoring some variables, which is okay in many cases, depending on their relevance.

But just as I can properly say that the Republicans are probably going to win a majority in at least one of the chambers of Congress on November 8, I can say that the chances that the Republicans win a majority in at least one of the chambers of Congress on November 8 is >0.5. The two ways of saying it mean the same. The use of numbers allow us for more nuances, since we have more options that just the ordinary English 'probably', 'very probably', 'extremely probably', and we run out of options bar synonyms - or perhaps a few more, but very few -, whereas with numbers I can say for example >0.7, <0.9, etc. As I mentioned earlier, there is a difficulty with the precision of that since our actual assignments probably (!) aren't as precise as assignments with any number of decimals, but even 1 decimal is an improvement. And two decimals are commonly used in prediction markets and work just fine, since that would be the same as the more colloquial % notation that you used above.

Now, just as one can properly make such intuitive assignments without using Bayes's theorem or any mathematical machinery consciously, one can make the intuitive assessments and then use Bayes to get conditional probabilities of some events - and that's okay, as Bayes's theorem and generally the mathematical theory is a good model of the epistemic probability, so both ways of doing it are right.
Don2 (Don1 Revised) said:
That's kind of ridiculous. It would be like me saying my probabilistic assessment based on all available evidence is intuitively that the P(X)<.25.
And of course, that would not only not be ridiculous, but just fine and ordinary. Just as I can properly say that my probabilistic assessment based on all available evidence is that something is improbable, or very improbable, I can better convey my assessment by saying it's <0.25, or <0.1. It's just a way of using a richer language, with more words, in order to transmit more info, rather than just 'improbable', 'very improbable', 'extremely improbable' or synonyms.

By the way, are you familiar with prediction markets?

#### Don2 (Don1 Revised)

##### Contributor
Don2 (Don1 Revised) said:
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one.
No, that is false.

No, it is true. The probability of a past event of the kind under discussion is 0% or 100%.

First, we are talking about epistemic probability of course, and of course it's neither 0 nor 1 (or 0%/100% in your notation).

And when you ignore facts, cherry pick stringency, handwave a bunch of 1's, and don't mention some unknowns, epistemic probability is pseudoscience.

Second, with your criterion, if the universe happens to be deterministic, then the future probability would also be either 0 or 1.

That could be, but it's also irrelevant and impractical.

Don2 (Don1 Revised) said:
If you were actually able to look at each and every involved issue and show the probabilities, you might have some kind of point, but your assessment of one or the other using Bayesian statistics to derive a mathematical statement that P(X)>.5 is pseudoscience because you're just listing out some incomplete things, some cherry-picked stringencies by Wade, some not included, and some unknown and then saying "but I'm allowed to do it because it's intuition!"
We humans make intuitive probabilistic assessments all the time. Mathematical probability is a model of that probability. It allows sometimes for greater precision, at the cost of ignoring some variables, which is okay in many cases, depending on their relevance.

But just as I can properly say that the Republicans are probably going to win a majority in at least one of the chambers of Congress on November 8, I can say that the chances that the Republicans win a majority in at least one of the chambers of Congress on November 8 is >0.5. The two ways of saying it mean the same. The use of numbers allow us for more nuances, since we have more options that just the ordinary English 'probably', 'very probably', 'extremely probably', and we run out of options bar synonyms - or perhaps a few more, but very few -, whereas with numbers I can say for example >0.7, <0.9, etc. As I mentioned earlier, there is a difficulty with the precision of that since our actual assignments probably (!) aren't as precise as assignments with any number of decimals, but even 1 decimal is an improvement. And two decimals are commonly used in prediction markets and work just fine, since that would be the same as the more colloquial % notation that you used above.

Now, just as one can properly make such intuitive assignments without using Bayes's theorem or any mathematical machinery consciously, one can make the intuitive assessments and then use Bayes to get conditional probabilities of some events - and that's okay, as Bayes's theorem and generally the mathematical theory is a good model of the epistemic probability, so both ways of doing it are right.

Here is what I wrote: "your assessment of one or the other using Bayesian statistics to derive a mathematical statement that P(X)>.5 is pseudoscience because you're just listing out some incomplete things, some cherry-picked stringencies by Wade, some not included, and some unknown and then saying 'but I'm allowed to do it because it's intuition!' "

Don2 (Don1 Revised) said:
That's kind of ridiculous. It would be like me saying my probabilistic assessment based on all available evidence is intuitively that the P(X)<.25.
And of course, that would not only not be ridiculous, but just fine and ordinary. Just as I can properly say that my probabilistic assessment based on all available evidence is that something is improbable, or very improbable, I can better convey my assessment by saying it's <0.25, or <0.1. It's just a way of using a richer language, with more words, in order to transmit more info, rather than just 'improbable', 'very improbable', 'extremely improbable' or synonyms.

And it's still bs.

#### laughing dog

##### Contributor
Don2 (Don1 Revised) said:
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one.
No, that is false.
First, we are talking about epistemic probability of course, and of course it's neither 0 nor 1 (or 0%/100% in your notation).

Second, with your criterion, if the universe happens to be deterministic, then the future probability would also be either 0 or 1.
That is false, because the future has yet to occur, and Don explicitly is referring to the true probability of a past event, not something in the future.

#### Angra Mainyu

##### Veteran Member
Don2 (Don1 Revised) said:
No, it is true. The probability of a past event of the kind under discussion is 0% or 100%.
No, it is false. The probability that yesterday (UTC), Trump ate a banana (for example) is neither. I would put it at somewhat less than 0.1, but more than 0.05. In this case, my info is general statistics about US banana consumption, and have no further info on the matter on Trump. But sometimes, one does not have statistics, yet that does not prevent proper assessments. The probability that Trump yesterday said "Putin" is neither 0 nor 1; I would go with 0.9 given current events.
Don2 (Don1 Revised) said:
And when you ignore facts, cherry pick stringency, handwave a bunch of 1's, and don't mention some unknowns, epistemic probability is pseudoscience.
No. Let us separate: if one deliberately ignores facts or cherry picks, that would be improper. Unknowns, on the other hand, are always there.

Don2 (Don1 Revised) said:
That's right, but it's also irrelevant and impractical.
Impractical? Well, if you go by that, assigning 0 or 1 to past events is just as impractical.
But it was a reductio, and you bit the bullet.

Don2 (Don1 Revised) said:
But B20 is not cherry picking, and as for the rest, of course information is always incomplete, there always are unknowns etc. That doesn't make epistemic probability pseudo science. It's still just ordinary human behavior.
Don2 (Don1 Revised) said:
And it's still bs.
No, it's still proper and ordinary, daily human behavior. Instead of limiting my expression of my probabilistic assessments to 'probable', 'improbable' and the qualifiers 'very', 'extremely', 'rather', and synonyms (and maybe one or two more), I use some other words, like '0.5', '0.8', etc.

By the way, are you familiar with prediction markets?

#### Angra Mainyu

##### Veteran Member
Don2 (Don1 Revised) said:
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one.
No, that is false.
First, we are talking about epistemic probability of course, and of course it's neither 0 nor 1 (or 0%/100% in your notation).

Second, with your criterion, if the universe happens to be deterministic, then the future probability would also be either 0 or 1.
That is false, because the future has yet to occur, and Don explicitly is referring to the true probability of a past event, not something in the future.
Funnily, Don bit the bullet on that one, so I suggest you take it up with him.

#### laughing dog

##### Contributor
Don2 (Don1 Revised) said:
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one.
No, that is false.
First, we are talking about epistemic probability of course, and of course it's neither 0 nor 1 (or 0%/100% in your notation).

Second, with your criterion, if the universe happens to be deterministic, then the future probability would also be either 0 or 1.
That is false, because the future has yet to occur, and Don explicitly is referring to the true probability of a past event, not something in the future.
Funnily, Don bit the bullet on that one, so I suggest you take it up with him.
You took it up with him as well, and you took it up poorly.

#### Don2 (Don1 Revised)

##### Contributor
Interesting article by Dembski saying it's intelligent design style inferences:
Dr Dembski said:
My interest here is simply in the logic of covering up design inferences and how the reports (rumors) I’m hearing about China trying to cover up a design inference vis-a-vis the human origin of the coronavirus may illustrate that logic. Briefly, what I’m seeing is a pre-theoretic design inference without laying out the full specifications and (im)probabilities that must be made explicit in any full-fledged inference of this sort. Thus, in yesterday’s Wall Street Journal, medical doctor Stephen Quay and Berkeley physics professor Richard Muller write:

The possibility that the pandemic began with an escape from the Wuhan Institute of Virology is attracting fresh attention… [T]he most compelling reason to favor the lab leak hypothesis is firmly based in science. In particular, consider the genetic fingerprint of CoV-2, the novel coronavirus responsible for the disease Covid-19. In gain-of-function research, a microbiologist can increase the lethality of a coronavirus enormously by splicing a special sequence into its genome at a prime location… n the entire class of coronaviruses that includes CoV-2, the CGG-CGG combination has never been found naturally. That means the common method of viruses picking up new skills, called recombination, cannot operate here. A virus simply cannot pick up a sequence from another virus if that sequence isn’t present in any other virus… Although the double CGG is suppressed naturally, the opposite is true in laboratory work… Now the damning fact. It was this exact sequence that appears in CoV-2. Proponents of zoonotic origin must explain why the novel coronavirus, when it mutated or recombined, happened to pick its least favorite combination, the double CGG. Why did it replicate the choice the lab’s gain-of-function researchers would have made?

If this is not a full design inference in the sense of my Cambridge University Press monograph, it’s close.

#### Bomb#20

##### Contributor
Clearly, you're not an expert neither. You are describing some math, but you are not giving any background basis for what the inputs should be, because you don't have it. Don is right about it sounding like ID, this is no different than when Michael Behe comes up with some probability that proves evolution couldn't produce some protein. Just like Behe does, your .05 x .05 assumes simultaneous mutation. And a 1/400 number is hardly a big hurdle anyway when you consider a single covid infection produces billions of copies.
I.e., you guys and Don didn't understand the argument.
Nobody misunderstood your argument, they just disagreed with it as well as the facts and reasoning you used, obtained from Wade.
Bomb#20 said:
Nobody is claiming SARS1.9 couldn't have naturally evolved a furin cleavage site.
But you are claiming it is a very rare thing which is a problem for at least two reasons. (1) Your information from Wade was wrong--it's common
Um, where am I supposed to have claimed evolving a furin cleavage site is rare? How would it even make sense for me to say such a thing? Rare compared to what? I was making an argument about relative probabilities.

and (2) even if it were not common, that's the point of natural evolution...in a world of things it does, the results are often an uncommon or improbable thing. Creationists often argue how unique or specific the thing is, rather than taking an outside view that there are a gazillion improbable things and one ends up as a result. It's like if you buy a lottery ticket and use the machine to select the numbers and they come back with some combination--any combination--you can't then say what were the odds of that?! I understand it's more nuanced in a sense, but in a broader view of not merely the lottery ticket, you also have a hundred people each choosing a lottery ticket and you are ignoring all of them except 1, saying, "see look?! this one here has some consecutive numbers! That was most likely intelligently designed."
Thank you -- you have just proven beyond a shadow of a doubt that you didn't understand the argument. The other lottery ticket numbers weren't any more likely than the number you got.

It's very simple. When there were four hundred ways for something to happen and the one that happened is one that had a 1/400 chance, that's not an improbable coincidence. When there were thirty-six ways for something to happen and the one that happened is one that had a 1/400 chance, that is an improbable coincidence.

There are many other features that nature might evolve that would also be suspicious to conspiracy theorists. The probability is high that at least one naturally evolved feature would be cause to start screaming about intelligently designed inserts into the genome.
Every time you guys say the word "conspiracy" you show you aren't interested in a good-faith discussion.

Bomb#20 said:
This isn't a Behe argument; I'm not assuming simultaneous mutation; and the number of copies/rolls of the dice is irrelevant.
No, the number of rolls of the dice is relevant, but perhaps not in the exact way you mean.
I already explained why it isn't. You have failed to show any way the number of rolls of the dice is relevant.

Bomb#20 said:
If one of the natural origin scenarios is what happened then the sequence would probably have been different from what it's observed to be;
Yes, odds are that you can find an evolved feature out of many evolved features and say, "the odds of that one thing happening that way are too rare."
But I didn't say the odds are too rare. I said the odds made it rational to reduce one's estimate of the likelihood of one scenario compared to another. "Too rare" is a property not of a single observation such as this one, but of all the observations collectively in combination with one's subjective prior estimate of the relative probabilities. If you don't grok that, go read an article on probabilistic reasoning.

Bomb#20 said:
therefore the observation that it's CGG-CGG makes those scenarios less likely.
I disagree with the philosophy. The true probability of a past event is either 0% or 100%, we just don't know which one.
So let me get this straight. Are you telling us that when Alan Turing's team broke the Germans' Enigma code, they must have actually done it using Turing's preternatural oracular connection to the knowledge of the gods, and not by means of all the mathematical calculations they made with their Bombe machine, since those mathematical calculations were actually all wrong, because those calculations were all based on the false assumption that the probability of whether this letter in the code represented E was the frequency of E in German, when in fact since the intercepted message had already been written and encoded, so whether that letter had been encoded from an E in the original message was a past event, therefore its true probability was either 0% or 100% but certainly not 17%?

If you were actually able to look at each and every involved issue and show the probabilities, you might have some kind of point, but your assessment of one or the other using Bayesian statistics to derive a mathematical statement that P(X)>.5 is pseudoscience because you're just listing out some incomplete things, some cherry-picked stringencies by Wade, some not included, and some unknown and then saying "but I'm allowed to do it because it's intuition!" That's kind of ridiculous. It would be like me saying my probabilistic assessment based on all available evidence is intuitively that the P(X)<.25.
What's ridiculous about that? Go for it! This is why people make different bets about the future: different intuitions and familiarity with different subsets of the evidence. You're not irrational to judge P(X)<.25.

Since we're talking about CGG, it should be stated that CGG is common in codon usage.
In mammals. It's rare in coronaviruses. Viruses normally recombine with RNA from other viruses, not from the host organism.

So IF it's naturally evolved OR recombinant from an intermediate host, then all that is required is the 100% probable past event that such intermediate host had CGG as common codon. We don't actually know the codon usage in most organisms. We don't know if mink has CGG as a common usage or not. However, we DO know it's common with hamster and that hamster could spread the virus just like mink.
Well, it could if the virus had a furin cleavage site. But if you postulate that SARS2 got its furin cleavage site by splicing it in from hamster RNA then you're postulating that the hamster had an infection of some pre-mutation SARS1.9 with no furin cleavage site. We don't know that hamsters can spread SARS1.9 just like mink can spread SARS2. The hamster hypothesis has a chicken-and-egg problem. Not impossible, but it would be another improbable coincidence.

The animals we know can spread the virus got it from humans.

What I can also tell you is that the 12nt sequence is in the hamster genome. Just do a BLAST to confirm. Here is top hit:
Cricetulus griseus strain 17A/GY unplaced genomic scaffold, alternate assembly CriGri-PICRH-1.0 unplaced_scaffold_55, whole genome shotgun sequence
Query  1      CCTCGGCGGGCA  12
||||||||||||
Sbjct  85478  CCTCGGCGGGCA  85489

The point isn't to promote a hamster hypothesis, just to show it's more common than let on.

The 12nt sequence also hits to American mink. Here is top hit: ...
And here's pangolin: ...
And all that would amount to a good argument if there were any evidence that in 2019 SARS2 was endemic in the local Chinese wild hamster/pangolin/American mink population. But guessing a virus jumped to a human from an animal that that virus did not normally infect is postulating an improbable coincidence.

Last edited:

#### Metaphor

##### Zarobljenik u hrastu
Banned
US Senate Committee concludes likely lab origin for COVID-19

From the conclusion:

Based on the analysis of the publicly available information, it appears reasonable to conclude that
the COVID-19 pandemic was, more likely than not, the result of a research-related incident. New
information, made publicly available and independently verifiable, could change this assessment. However,
the hypothesis of a natural zoonotic origin no longer deserves the benefit of the doubt, or the presumption
of accuracy.

#### Rhea

##### Cyborg with a Tiara
Staff member
Make note that this is from the “HELP Minority Committee,” note specifically “minority”. Meaning it is from the Republican caucus. How many of them? Well,

The report is signed by only one person, Senator Burr. No others put their name on the document.

The members of the Senate Committee on Health Education, Labor and Pensions are a large group. But they do NOT put their pen to this. And of the “Minority Oversight Staff” listed as authors? One. Just one. Who might have signed but didn’t?

MINORITY BY RANK

Richard Burr (NC) Rand Paul, M.D. (KY) Susan Collins (ME) Bill Cassidy, M.D. (LA) Lisa Murkowski (AK) Mike Braun (IN) Roger Marshall, M.D. (KS) Tim Scott (SC) Mitt Romney (UT) Tommy Tuberville (AL) Jerry Moran (Kansas)

Couldn’t even get Tommy Tuberville’s signature?

In the end, he very clearly states his “conclusion” as Just Asking Questions, (you edited that part out of your quote) and not as having demonstrated anything. He’s just maintaining public agitprop.

A prudent person would NOT take his conclusion as difinitive And certainly not as a definitive statement of our government.

#### Loren Pechtel

##### Super Moderator
Staff member
US Senate Committee concludes likely lab origin for COVID-19

From the conclusion:

Based on the analysis of the publicly available information, it appears reasonable to conclude that
the COVID-19 pandemic was, more likely than not, the result of a research-related incident. New
information, made publicly available and independently verifiable, could change this assessment. However,
the hypothesis of a natural zoonotic origin no longer deserves the benefit of the doubt, or the presumption
of accuracy.
This is the Republican report, not the real report.

#### Metaphor

##### Zarobljenik u hrastu
Banned
US Senate Committee concludes likely lab origin for COVID-19

From the conclusion:

Based on the analysis of the publicly available information, it appears reasonable to conclude that
the COVID-19 pandemic was, more likely than not, the result of a research-related incident. New
information, made publicly available and independently verifiable, could change this assessment. However,
the hypothesis of a natural zoonotic origin no longer deserves the benefit of the doubt, or the presumption
of accuracy.
This is the Republican report, not the real report.
What's the "real" report?