This seems like a mischaracterization. It also sounds like the drop of water that marveled how God made a puddle just the right shape to hold it.
The disease reservoir in that area is probably holding dozens if not hundreds of dangerous viruses. Perhaps not specifically adapted for indoors transmission - maybe one is transmitted by touch, maybe one has a very long incubation period. The odds of some sort of deadly disease coming from this population is high. That’s why they located an entire lab there. It’s happened before and it could happen again tomorrow.
I also don’t understand your leaning on the idea that the virus left no trace of how it got there. Viruses typically don’t log their comings and goings. The only information about them is in the RNA, and there’s been quite a bit of comparative analysis on other known samples. They haven’t reached a conclusion yet IIRC, but their failure to do so doesn’t automatically mean someone’s hiding it.
I think you misunderstand exactly what P(N) is. P(N) is not the probability that the virus (or its immediate precursors) evolved in the wild. That we know for sure is the case, that probability is essentially 1.
P(N) is the probability that there is no causal connection to the Wuhan lab whatsoever. So 1-P(N) would include, for example:
The virus evolved in the wild, then a Wuhan-based researcher went out into the field, obtained samples somewhere and brought them back to the lab, etc.
The virus evolved in the wild, then a Wuhan-based researcher was out in the field and became infected with SARS-CoV-2 in the wild and (possibly asymptomatically and unknowingly) brought the disease back to Wuhan, creating the causal connection between the Wuhan lab and the Wuhan outbreak perhaps without the virus ever being in the lab.
And there is a zoo on the plain of horses, but it isn’t right next door, it is 30km away across a river. That’s the distance between the lab and the wet market.
We know SARS-Cov-2 made its way to the wet market, which was the first noticed super spreader event, either directly from patient zero, or through a chain of infections from patient zero. The question we seem to be debating, is whether patient zero was associated with the lab, or was just the unlucky -otic part of zoonotic transmission (-otic means human in Greek, right?) someplace, probably in southern China.
So, what’s the prior probability of this, which certainly is not an original hypothesis. Patient zero is in the wilds, catching wild animals, perhaps bats or pangolins, or just playing in their poop or whatever. Patient zero transmits SARS-Cov-2 along with some wild animals to somebody else, who perhaps does the same thing again. Eventually a person carrying SARS-Cov-2 and those caught wild animals goes to a place in the nearby metropolis to sell the wild animals. A sneeze and a cough later and we have a world wide pandemic.
Right, this seems the most likely causal connection to the Wuhan. Not some sinister and bizarre experimental program, just somebody connected with the Wuhan lab doing their job as a coronavirus researcher out in the field and either collecting a sample that later infects someone else in Wuhan or catching an infection and becoming patient zero themselves and bringing it to Wuhan.
Yes, and I see that as possible, if not a likely. What I see as more likely, is that somebody out in the field collecting animals which were bound for the wet market, was responsible (maybe through an infection chain) of delivering the virus to the wet market.
But it is not more likely when you take into account the additional information that this was the wet market in Wuhan. Bayes again. There are many farmers and many possible markets where a random farmer could have brought the virus. There is only one Wuhan.
For those of you who still don’t seem to be able to get to grips with how the fact that the outbreak in Wuhan changes the posterior probabilities here…
What would you think if next week there was another first major outbreak of an unrelated coronavirus also in Wuhan? Would you think it was probably a coincidence?
It’s not the only wet market in Wuhan, it’s a massive city; they’re are many. It just happened to show up at this one, 30km and the other side of the river from the pathogen lab.
I’m arguing that as the primary city of the region it’s wrong to ignore the Bayesian probability that the virus would travel there anyway. People have lots of reasons to go to major cities. It happens all of the time. If you are 100 miles away from a major city, how many degrees of separation are you from somebody who will be going there in the next week?
Additionally, the prior probability is 1 that wild animals sold at that wet market were brought to the city from the wild by people.
So, where is Bayes for one of a few researchers collecting wild samples, taking precautions, bringing the virus back to the lab, where precautions (even if flawed) are taken, then having the virus infect a person, who directly or through a chain of infections takes it to a wet market 30kms away?
Compare that to the probability that one of the many hunters and trappers, who don’t take pathogen precautions when handling animals, picked up they virus and either directly or through a chain of infection brought the virus to the place the animals they trapped were going?
That is not a remotely coherent probability analysis.
But what proportion of ALL trappers within the range of the relevant species are located specifically near Wuhan? That’s what’s relevant.
We’re just going over the same ground again and again. The point is very simple. If the Wuhan lab were not implicated in any way, the denominator of the probability of this happening by chance in Wuhan is ALL the possible places in the entire region (including neighboring countries within the range of the relevant species) where this could have happened. What proportion of ALL the possible places that this could have happened does Wuhan represent?
You really don’t need math for this, it’s just common sense.
Answer this question, please:
So, obviously it’s a rhetorical question. You would know for sure it wasn’t a coincidence. Because that would be the THIRD thing associated specifically with Wuhan (the lab location and two outbreaks), so it’s vanishingly unlikely to occur by chance - there must be some specific causal connection to Wuhan.
Yet you insist that TWO things associated specifically with Wuhan - the lab being located there and the first outbreak occuring there - has no probabilisitic significance.
It doesn’t have to be perfectly adapted. It could have jumped and circulated in human populations for years before evolving in humans for better transmission and better binding to cellular receptors.
You are given a bag containing ten identical-looking dice. You are told that there are nine fair dice, that will roll 1 through 6 with equal probability; and one loaded dice that will always roll 6. You are asked to pick one and then roll it repeatedly to try to identify whether it’s fair or loaded.
So, the prior probability (no rolls) is obviously 10% that it’s the loaded dice.
Your first roll is a 6.
It could still be either, of course. But is the probability that it’s a loaded dice still 10%, or is it higher?
I would say that China has a big problem with habitat reduction and humans encroaching on wild animals. Scientists have been warning about this for decades. China’s economic growth has made this a big problem. As humans and animals come into conflict, the chances of a zoonotic transfer greatly increase.
And, this is the second outbreak. We’ve already seen SARS-1 come out of China. It didn’t appear anywhere near Wuhan. It was first identified in Foshan municipality, a city of about 7.5 million. So another case of a virus suspected of being reservoired in bats first being identified in a major city in China.
Why didn’t the super spreader event start in a cafe or bar near the lab if that was it’s origin? Why not to some random place nearby where lab workers visit? Why did it travel 19 miles to a place frequented by people who interact with wild animals?
If the outbreaks were in the immediate vicinity of the lab, or at the workplaces of lab workers’ families, or some direct chain like that, then the probability of the lab being involved goes way up. If further research turns up a defrosted vial in the lab with residue genetically identical to SARS-Cov-2 with a label saying “collected July, 2016”, then that’s a smoking gun.
Finally, it is important to remember that Wuhan (just like Fushan with SARS-1) is just where it was first identified, not where it first occurred. Even a decade after SARS-1 there wasn’t a good handle of where it came from. We may never know.
I’m one who doesn’t have knowledge (and thus appreciation) of exactly how common such events are in that area. How common are they? By common do you mean once every couple decades, like with SARS and its sequel? Or do you mean like once a month?
No doubt that is what must’ve happened. We currently don’t have any record or evidence of that prior jumping and intermediate steps and so until we do the only only honest approach is to keep all options on the table.
Yes, that’s the question. There were plenty of very early cases that had no connection to the wet market and this has been known from very early on.
What we don’t have is an evidencial pathway for how it got there.
Purely natural mechanisms, trade and travel will have put potentially dangerous viruses into that Wuhan area, that’s almost certain.
However, The Wuhan lab will also have put potentially dangerous viruses into the Wuhan area. And no “almost” about that, that fact has been admitted to by the lab leaders.
So we are left trying to evaluate which is more likely. The longer chain of Human/animal interaction possibly bringing the virus to Wuhan purely by natural means and chance, spreading to an unknown degree in that area and then superspreading at the wet market.
Or starting with the known location of a potenially problematic virus in the Wuhan lab and having that make its way to the Wuhan population with the same spreading/superspreading events.
I’m not in a position to make that evaluation and have seen nothing in the official responses to make me bet my house on one or the other.
Yes, think of the pandemics (or pathogens of concern) that have already come out of that area. They emerge every few years because the pace of evolutionary change is very fast for small organisms. Since we can assume that like most wildlife, most of these viruses never come into contact with humans, that there are always a few very nasty ones always waiting in the wings that never have a chance to get established in the human population.
I like the fact that we’re starting to rule out the more dramatic scenarios (engineered viruses, a failure of sophisticated lab protocols). The more we do that, the more the characterization begins to look like “a mistake was made by a person who routinely encounters wild specimens of bats and pangolins.” Then we can understand more fully why lab techs aren’t at a substantially elevated chance of encountering a particular pathogen.
If we consider that most people in a lab actually don’t collect or handle living samples (probably in the low double digits), then the lab’s potential to contribute seems much lower. Certainly much lower compared to the space of hunters, farmers, zoo workers, wet-market butchers, and any number of other people in that area who cross paths with those animals.
If we consider that animals don’t advertise what viruses they harbor, then we can also conclude that there’s no reason a lab tech who collects specimens of bats or pangolins should be at any special risk to harvest Sars-Cov-2 as opposed to any of the thousands of other viruses that circulate out there, even if they were specifically looking for that virus.
According to one of the articles I posted, Shi Zhengli’s group does collect samples from bats, other mammals, and they receive patient blood samples. However, they are much less likely to catch a coronavirus than a farm worker, wet-market butcher, or zoo worker because they would be taking precautions in collection. After they collect (according to them), they mainly just extract genomes for sequencing. Not a lot of cell culture going on.
I think people are not appreciating how many coronaviruses are out there in many different species of mammals.
Nope, you still don’t understand how to think about probabilities. We’re not lacking good theories of how the virus could have spread to people without lab involvement. Expounding them yet again adds nothing. We all agree the PRIOR probability of lab involvement, if the outbreak had been in a random place other than Wuhan, is extremely low - because nature is a huge place compared to a lab, because the number of lab workers is small as a proportion of all the other people coming into contact with animals.
But all these other people are spread across a very large region, and it could have happened anywhere to any of them. What we’re lacking is any good explanation other than lab involvement that explains the unlikely fact that the outbreak took place specifically in the place where the lab is located, in Wuhan. Nothing you have said here speaks to that.
Again, you don’t understand how to think about probabilities. We’re know that SARS-CoV-2 did somehow cross to some human. The large number of other viruses might have spread is irrelevant. The numbers that did not come up on prior dice rolls don’t matter unless different groups of people were rolling different dice. SARS-CoV-2 was just one of thousands of viruses harbored by animals, but that’s true for a farmer just as much as it’s true for a lab worker. They were rolling the same dice in that respect, so when we know this virus did cross, this does not speak to the probability of exactly who it crossed to.