DawgTalkers.net Forums DawgTalk Palus Politicus Scientific American

What the hell are you talking about?

This person is no longer in Hong Kong. And the vast majority of immunology researchers in the world are not located in China.... so I don't really get your point?

Yes, but they're all controlled by the Chinese government.

Just like NASA spends all their money paying off hundreds of thousands of scientists, pilots, astronauts and meteorologists to say the Earth is round....

Now you get it!

So you guys are saying there is no way she could be right?

Your Chinese masters will be happy with you.

There's an overwhelming consensus in the scientific community that says she's wrong. And they're not from China comrade.

Who cares. The "Science" folk are every bit as bad as the "religion" folk.

I swear these same science folk were telling us masks don't work. Trump isn't godly and Biden doesn't have anything to do with science. Just utter nonsensical partisan posturing.

There is a difference between proving something is impossible and believing something isn't plausible. Belief isn't science.

Assuming everyone would start the same way if trying to engineer a virus is a logical fallacy. People think differently and have different approaches. You don't have scientific breakthroughs by doing things the same way everyone else is.

The tens of millions and monkeys with typewriters example seems daunting. It's less daunting when you take advances in technology, like supercomputers, into account. link

None of the scientists mentioned in these articles had ever thought to use a computer... you should write to them! (if they can even open their e-mails)...

Have you been taking straw man lessons from Pit?

There was an actual point there.

When Robert Garry said "this is impossible" he was obviously taking into account the idea that people might try to solve these problems on supercomputers. Even then, it is impossible.

EDIT: In fact, this is exactly what Kristian Andersen (the lead author of the study) works on: https://andersen-lab.com/

"All the scientists but one" is suddenly a straw man argument. How do they manage to make this crap up?

Implausible is not impossible. Your certainty is troubling. It's not science. It's belief.

This seeming Science as religion/Scientist as High Priest trend makes no sense to me. They said it can't be done, it must be impossible. They can't conceive of it, it must not be possible.

"The more you know, the more you realize you don't know" rings true to me. Unfortunately, in the publish or die "scientific"/academic community, this doesn't seem to be realized or admitted any more.

You asked, and I quote, "Can you imagine how many there are that disagree at University of Hong Kong?"

I answered your question. I guess you forgot your own question. No one inside of China will disagree with Chinese dogma unless they've defected, or want a permanent vacation to a camp.

It is "possible" that when I throw a ball at a wall - the ball will quantum teleport through the wall and emerge unaffected on the other side. However, when I do science - I don't give much weight to such possibilities.

We are all Bayesians at heart (and we should be!)
https://en.wikipedia.org/wiki/Bayesian_statistics

If top experts (in a subfield) are saying there is a 1 in a billion chance that some event can happen -- then we should recalibrate our priors based on this information and make more informed judgements.

That is what science is -- and that is what we are all (most of us) doing.

Li-Ming Yan fled China within the last six months and has been mostly in hiding within the U.S. Supposedly the ROL Foundation helped her escape China.
She claims to have worked at the laboratory.
She appears legit. And whether she is legit or just a political stooge, the findings in the paper are legit and solid. Its not like the paper was written in a vacuum. There are many scientists analyzing the same genetic data and coming to the same conclusions. She cites 111 sources in her paper that are mostly works authored by highly qualified biological scientists. The evidence that she provides in here paper is 100% dead on.

A lot of mainstream journalists will claim that there is some scientific consensus on this and like to cite the paper "The proximal origin of SARS-CoV-2" which was published on March 17. This is the paper that argues for the connection to the pangolin as an intermediate host. This paper actually got me interested in the topic, because it is so full of bad conjecture, bad data pedigree, and misinformation that you wonder how it ever got published. I am by no means an expert, but I know enough to look at genetic sequences and amino acid sequences and draw some conclusions, and when I read that paper, I couldn't believe how they just glossed over glaring evidence that didn't support their conclusion.
The paper has been thoroughly debunked at this point by scientists.

Robert Garry is not being forthright. The whole basis of his claims is that because the binding site of the spike protein wouldn't have been predicted by a computer simulation, it could not be man-made.
We have been doing insertions into proteins for longer than most scientists have had access to supercomputers.
Besides, the analysis of the binding site shows it is a near perfect match for SARS on the amino acids that are active in creating the binding.

If you were doing gain-of-function research that sought to study what would happen if a SARS binding site were to recombine with other coronaviruses, this is what you would want to create. You wouldn't use a computer simulation to come up with some arbitrary binding domain that does not exist in nature.

Oh look, another extreme/ridiculous example that no one on the other side brought up to try to make your argument feel sane in comparison.




Biological warfare is a reality. Fact. People are doing research on genetically modifying viruses. Fact.

Yet, somehow you can't imagine how this could possibly have been engineered because it wasn't done the way other scientists would have done it?

It's a coincidence that COVID-19 appears to have originated near a research facility in Wuhan that was working with 3 similar strains of coronavirus?

I'm also not sure you understand the computing power of modern supercomputers.

Sidenote: it's kind of hard to take Andersen Lab serious when a job title for a good chunk of their people is "cat herder" and the options/buttons at the top of their website are Heroes, Mysteries, Poetry, Gossip, and Tales.

Edit: Want to reiterate I'm not saying it definitely was man-made. Just refuting the assertion that it couldn't have been.

Please post -- any of them?



I agree - anybody can cite a paper... For example, citation 4 is the Andersen study that disagrees with her (that is the right thing to do! You should always cite relevant papers regardless of whether the conclusions agree)



According to... you now?



Again... please list... any of them.

Yes - probably...



I have worked on several of them...



Don't attack the messenger - remember?

That's why I put it in a sidenote. *shrug*

edit: I'll also add that working on something doesn't necessarily mean understanding. Out of curiosity, When were the supercomputers you worked with built?

Edit 2: look at the specs on the Sunway TaihuLight. They're pretty insane, and this list is from 2017. China isn't playing around when it comes to supercomputers.

link

That is entirely a fair point. I do not build them - and am (due to specialty) somewhat behind on some of the new advancements in GPU acceleration -- because they are not often applicable to what I do.

They are modern machines though...

Current list is here:

https://www.top500.org/lists/top500/list/2020/06/

With all those crazy numbers, I'm starting to wonder if the constraint might be on the programming side. We have lots of raw computing power, but figuring out how best to utilize it for different applications is probably still being figured out.

So basically I think the hardware is in place that could make the million monkey "impossibility" possible. The question is the software-side.

Oh - most definitely... one example is that 2020 Chess Software on a 2005 Computer beats 2005 Chess Software on a 2020 computer.

But you can't brute force this problem, which is what Robert Garry is saying. The receptor-binding domain of Sars-Cov-2 is unique -- and contains a sequence of 193 amino acids. Single point mutations likely ruin the RBD.

There are 20^193 = 1.2e251 possibilities. No machines can even get close to that. (n.b. obviously many of these possibilities probably work in an identical or similar way -- even nature can't make a mutation that is that rare.)

The only way is to have a lot of knowledge that lets you narrow down the possibilities. The argument of the Nature paper is that we don't have enough knowledge to get us down to the 1e10 or so possibilities that are computationally testable.

Lol, still not a fan of statements that include things like "only way" that imply unsubstantiated certainty.

Another way of looking at the problem is trying to figure out how one would create such knowledge. Random "sampling"/ trial and error, pattern recognition software, a million other things I'm not thinking of or even aware of their existence. There could still be a lot of brute force not getting anywhere, but it'd only need to get "lucky" once. Honestly, I'm kind of still trying to wrap my head around "93 quadrillion calculations a second." link

If someone were using supercomputers to work on biological weapons/bio weapons defense, they'd probably keep any software they developed proprietary and all the related work would be highly classified. When it comes to programming, sometimes advances come from thinking about a different direction to come at a problem.

I think you are just passing off the problem of "scientist don't understand how binding domains interact" problem off onto computers, rather than onto people.

Whether you are using deep learning, some neural network, etc. -- at some point you need to be able to get rid of 99.9999999999999999999999999999999999 999999999999999999999999999999999999999999999999999999999 9999999999999999999999999999999999999 999999999999999999999999999999999999 9999999999999999999999999999999999999999999999999999% or so of the possibilities to have a hope of calculating this. There is no way that we made AI that is good enough to do this, while also having no conception of how it is being done.

I can state with certainty that you can't test 10^250 possibilities with a classical computer. There aren't enough particles in the universe (only about 10^90).

I'm saying you don't need to test all those possibilities. Someone just had to come up with a new, novel way of narrowing them down.

Absolutely - that is what nobody knows how to do. Computers can't help much with this either.

I would be more than happy to answer your questions and go down the rabbit hole, but only if you are willing to be open minded and have not already decided to be set on a specific opinion, because it is a lot of work to go down that rabbit hole and discuss the somewhat complex data behind the analysis of the virus's origins.

I think I have hijacked this thread which was really about Scientific American's decision to endorse Biden. Should we continue this in the COVID thread or a thread of its own?

Go for it.

Or at least nobody is saying that they know how to do. If they figured it out while working on bio weapons, they probably wouldn't be broadcasting it. They also would probably be trying to discredit anyone who is claiming that they do.

But then, I'm not claiming a certainty, just a possibility.

I would be careful about investing too much time into an educational post. This board isn't too interested in learning. It's about furthering one's own specific beliefs.

I do admire your spirit. Maybe start off small and see if you can garner intelligent conversation. Call it off if the usual suspects shoot down your thoughts w/their usual garbage.

I'd love to read an intellectual conversation on the topic. I won't be able to participate because I don't have enough knowledge about the topic, but I would like to learn.

Maybe - I am generally of the opinion that classified labs are not so far ahead of the public sector as people seem to think.

There's a couple choice regimes where the best work is done in the classified sector - cryptography, radiation transport simulations (important for nukes), etc.

There are a couple choice regimes where business is ahead of everybody else: deep learning, data science, drug design, quantitative analysis, to name a few.

There are other regimes where the best research is done at universities: neurobiology, astrophysics, ultra-low temperature physics, etc.

But it's not like "all the best minds always work on classified problems" -- there's not some "Serenity-like" government program that pulls supergeniuses out of the school system at a young age and sets them to work on classified science. The people who work there are just like the rest of us.

It only takes one brilliant/twisted mind.

I had to get a series of Anthrax shots while in the military, and this was over a decade ago. (While doing WESTPAC cruises and tensions were high with China) The worry about biological weapons has been around for a while. Symposiums were held on the subject (I was volun-told to work security at one.)

"We are in the realm now where biological weapons are really becoming possible,” the official said. “People have talked about [gene editing in bioweapons] for 50 years. … It is not science fiction anymore. Literally in the last five years we’ve crossed that threshold." link

"Some observers argue that gene editing could make it easier to develop or use biological weapons clandestinely, thus reducing the risk of international disapprobation. But maintaining a secret biological weapons program has never been particularly difficult. The equipment and agents required also have legitimate uses, and the challenges of international oversight mean that the odds of getting caught are low. It is unlikely that new technologies would change this in any fundamental way." link
(Probably not the most solid sources, but I'm using Google)

The ideas fit with my military experience.

I'm sure (relatively, I'm using the expression figuratively) there are a multitude of classified feasibility studies.(based on the prevalence of feasibility studies and the prevalence of the topic of biological weapons)

First I would point you to https://www.biorxiv.org/content/10.1101/2020.05.01.073262v1
While they don't state it in the document, the authors are openly critical of the pangolin article. Their argument is that the virus does not show the usual pattern of frequent mutation after the transfer to humans. It was highly adapted to humans from the first detection. Dr. Alina Chan in particular has been publicly critical on her twitter feed
https://twitter.com/Ayjchan
She is a researcher at the Broad Institute of MIT and Harvard and also part of the team developing https://covidcg.org/, a tool for analyzing and tracking Sars-COV-2 genetic mutations.
A little bit more mainstream article on her efforts and opinions here:
https://www.bostonmagazine.com/news/2020/09/09/alina-chan-broad-institute-coronavirus/

Another pair of authors are Jonathan Latham, PhD and Allison Wilson, PhD. They are staff at the Bioscience Resource Project and hold doctorates in Virology and Genetics respectively.
Here is one of several of their articles on the origins of the virus.
https://www.independentsciencenews.org/health/the-case-is-building-that-covid-19-had-a-lab-origin/

This preprint by Ressana Segreto and Yuri Deigin, basically calls them out right in the abstract.
https://www.researchgate.net/publication...ust_be_censored
"Based on our experience in genetic manipulation we cannot exclude a synthetic origin of SARS-CoV-2 and we believe that this topic should not be censored. In our manuscript we suggest a possible experiment that could have originated SARS-CoV-2, known to be chimeric and characterized by a furin cleavage site, missing in other beta-coronaviruses of the same lineage. Moreover, we do a critical analysis of the paper of Andersen and colleagues published in Nature on the Proximal Origin of SARS-CoV-2. This paper is considered to prove that SARS-CoV-2 has a natural origin, but in our opinion it lacks scientific evidence. We do not want to accuse a specific research group, but raise attention of the scientific community on this topic."
Yuri also published a more mainstream article on Medium

https://medium.com/@yurideigin/lab-made-cov2-genealogy-through-the-lens-of-gain-of-function-research-f96dd7413748

And if you truly don't want to read, you can see Yuri's interview with Bret Weinstein on Youtube. Before becoming a full time Youtube intellectual, Bret was a professor of evolutionary biology and happened to specialize in Bats early in his career.
https://www.youtube.com/watch?v=q5SRrsr-Iug

Each of these draws on others work and as you check their sources and references, you will find that the list of conspiracy theorists is getting longer. Outside of China and groups that have collaborated directly with WIV on the gain of function research, I cannot find anyone of authority that supports the conclusions of Andersen's paper.


Yes, that is my personal opinion on that one.


Sure - obviously any of the sources above are relevant, but frankly, the best evidence that computer simulation of a binding site is not needed to create a laboratory made version of a virus is prior instances of this sort of genetic engineering without using a computer simulation to optimize binding.
The Baric Lab did this exact thing when the put the human SARS binding domain onto a Bat SARS-like variant.
https://www.pnas.org/content/105/50/19944
and the Wuhan Institute of Virology has created similar SARS chimeras:
https://www.nature.com/articles/nm.3985
https://pubmed.ncbi.nlm.nih.gov/18077725/

So it is hard to make the case that nobody would create a SARS based chimera that was not designed by a computer when multiple labs have already done it multiple times.

So now, I agree that these are (more) legit people.

First: It is worth noting that most of these experts you mention are very critical of the Yan analysis. e.g., Ayjchan has a long tweet-storm about this, a snippet is below.



also:




2.) Many of these "experts" are pretty questionable -- none are active virologists. They certainly are more expert than me -- and I don't think they are fake people.

From least, to most, legit:

Yuri Deigan is not a scientist, he has an MBA and runs some "fountain of youth" site. He is definitely pretty sketchy.
http://www.youthereum.io/#rec50875635


Ressana Segreto is a scientist -- but this paper on SARS is the only paper she has ever written on viruses. The entire rest of her publication record is on fungi and plants.

https://www.researchgate.net/profile/Rossana_Segreto

Jonathan Latham and Allison Wilson are also not virologists - nor are they currently active researchers. They currently run a website together that focuses on bioengineered crops:

https://bioscienceresource.org/

Jonathan Latham's background is somewhat confusing. He lists himself as a PhD, but he doesn't mention where that PhD is from or anybody he has ever worked with.

I think he might be this person? https://onlinelibrary.wiley.com/doi/pdf/10.1046/j.1365-313X.1997.11061273.x


Alina Chan seems to be a legit expert -- though her argument is not that Coronavirus was made in a lab. She is mostly arguing that we shouldn't censor such research (and I agree with that!). But that doesn't make it correct, or defensible based on the science.


Third - onto the actual science:

I don't think any of these papers make the point that you think they are making.



This is exactly what the Andersen et al. paper is talking about (they cite a more recent paper by Cui, but the idea is the same).

What the Baric group does is take the Sars-Cov-1 RBD, and splice it into an existing Bat-based Coronavirus, and show that the resultant mutant virus strain can infect human cells. The RBD of their new virus is identical to the Sars-Cov-1 RBD.

Nobody denies that this is possible -- but this is not what the Sars-Cov-2 RBD looks like. And I would agree -- if we found another virus in the wild (which was being studied at the Wuhan lab, for example) which had a RBD very similar to Covid-19, then this would radically change my prior regarding a laboratory origin.




This is again the same idea. They took the spike protein from a bat coronavirus (SHC014) and placed it into a version of the humnan-based SARS-Cov virus that has been optimized to also infect Mice. They then found that this new hybrid virus (with the new Spike protein) was still able to infect mice, and infect human cells in vivo.

But again! This is a simple splice where one section of the virus looks 100% like SHC014, and the other section looks 100% like mouse-adapted Sars-Cov.




This is similar to the previous studies, except that, from what I understand, the combined virus they made failed to really infect anything....


Reading through the arguments from Chan (which I think is really the only person here worth reading) -- she makes an argument that you could produce Sars-Cov-2 in a lab. I totally agree that the gene splicing necessary is not beyond modern labs.

But this doesn't answer the more fundamental question that Andersen brings up. There is no way to know that the Sars-Cov-2 RBD would be infectious -- from first principles. It is easy to take something that is known to be infectious and splice it into a different virus. It is very hard to design (from first principles) an RBD that is known to attack ACE2.

So let's see if we get this straight. It's highly implausible but not impossible.

And then you try and say I'm the one with the straw man posts?



You've been given the evidence of how the virus' construction is unlike anything science has ever seen developed by man. That only virus' created through nature display such traits. That isn't "a belief".

So, I was upfront in saying that Yan might be a political stooge in all this and acknowledge that she is really just compiling work from other scientists and selectively documenting only the evidence that supports one side.

You can question the competence of my sources, but at a certain point you have to judge them based on what their arguments and evidence and at a minimum, they collectively make an incredibly strong case for not taking the conclusions of Andersen et. al. at face value. In my opinion, the analysis of genome and peptide sequences is not a skill that only a virologist can understand. It is pretty common across all biological sciences and you don't need a PhD to understand the relationships between nucleotide sequences and peptide sequences. From my standpoint, Deigin, Latham, Weinstein, and Chan all demonstrate at least reasonable level of competence to be taken seriously. If there analysis is bad, then a "qualified" person should tell them why their data is bad, not just dismiss it publicly as nonsense without explanation. They should educate the public. We are talking about a virus that will claim the lives of at least a million people.

Springer Nature, through Scientific American is putting out information that is misleading to the public, claiming that the scientific community has reached a consensus on the virus's origins and labeling individuals that dare point out evidence that is contradictory as "conspiracy theorists". The fact of the matter is that the key paper that the general press leans on for the natural origins argument is built on data that cannot be verified because their is no physical evidence of the pangolin or RaTG13 strains. All they have is strings of letters in a database uploaded by the WIV with suspicious timing.

I am not saying definitively that this is a man-made virus, although it may come off that way, that is definitely not something that I am willing to commit to. We don't know and may never know the origins of the virus because the evidence is either hidden in nature or destroyed by a Chinese lab.
What I am saying is that Springer Nature and Scientific American are on one side of a political battle by appearing to be unwilling to publish data that supports evidence undermining a flawed natural origin hypothesis, while knowingly publishing a hypothesis supported primarily by unverifiable data from what would be the guilty party in a lab creation or lab release scenario. They then report on this hypothesis as if it has consensus support from the scientific community, which it does not. I fully acknowledge that Yan is associated with and supported by a group that is against the CCP and has right wing political affiliations, which brings me back to the central theme, which is that I don't believe Scientific American is backing a political candidate out of sincere concern over how COVID-19 has been handled and how the President has misinformed the public. If that were the case, then they should be looking in the mirror, because they have been doing the exact same thing. I think they are engaged in a tit-for-tat political feud, not a forthright scientific debate. Their actions in all of this have been to the benefit of a foreign power that has a horrible human rights stance and has been far from transparent and honest with information related to the spread of the virus.

You can use this post as a prime example of what he's talking about.