> But Pfizer’s COVID-19 vaccine was developed in a few hours back in January 2020.
After which over 150 million people got sick and over 3 million people died. It seems possible that the regulatory rigidity responsible for the long roll out is a greater mortal threat than the virus itself. It's hard to imagine plausible scenarios in which a complete failure to regulate could have been more destructive. We have met the enemy and he is us. Apparently the most important technology that we lack to prevent a repeat isn't biotech, but the means to regulate the regulators.
The plan is in inject 6 Billion people with $vaccine. I have no medical expertise, but the idea of >3 million people dying from 6 Billion unregulated injections does not seem outlandish to me.
Edit: especially taking into account one of the main tasks of the regulators was to decide _which_ vaccine to choose. If I remember correctly, there were hundreds of vaccines in development. In hindsight we know which ones were the right choice, but that's the result of rigoros, regulated, time-consuming testing.
Let's also keep in mind that the timeline for approving the vaccine still pales in comparison to the timeline for manufacturing and distributing it.
Phase III trials for Pfizer took ~16 weeks. It has been 26 weeks since the US granted an EUA based on that trial. In the 26 weeks since then, ~14% of the world has been vaccinated.
And that's after we spent those 16 weeks of the trial building out manufacturing capacity.
We did skip a lot of steps on this vaccine. When we started mass administration of it, we had no idea whether it would prevent or reduce transmission or just symptomatic infection. We are still figuring out how long we expect immunity to last, whether we need boosters every six months or never.
Maybe once we have the technological capacity to manufacture and distribute 8 billion vaccine doses ~instantly it will make sense to skip the "does this even work?" step, but as long as that remains the bottleneck I have a hard time believing it.
You missed the does this have any long term effects on 6 billion people. Without long term testing we may find out in 5 years those who had the vaccine may suffer some side effects unknown today. More worrying is if in the future this delivery model becomes the norm where we roll out solutions before fully testing to save money.
In my opinion, that's the largest selling point of mRNA vaccines. In 4 weeks you can have the entire world's productivity capacity changed into whatever vaccine you deem urgent, and when it's not urgent, you can just store it in a computer's hard drive.
Once this is possible¹, we can ask governments to keep some extra capacity unused to respond to emergencies. What is different from any previous tech, where extra capacity would be useless nearly every time.
1 - AKA, when the patents expire. I vote for greatly rewarding the people that developed them and taking the patents away, but I'm sure that won't happen, so I hope we don't see any larger pandemics for 15 years.
I think back in April 2020, the NYT published a calculator where you could play around with the variables to see when a vaccine could be ready. The best you could do back then was mid 2021…
I think we have come a long way and pushed the boundaries substantially.
> Of late he feels like all the activity of himself and his peers is just playing the Science Game: varying some variable with infinite degrees of freedom and then throwing statistics at it until you get that reportable p-value and write up a narrative short story around it.
The classic "we achieve SOTA on ImageNet by using a novel training procedure" where they just played with learning rate schedules until they got 0.1% over previous SOTA.
To be fair to there are a lot of smells in DL papers, usually you can tell whether an approach is worth your time by looking at code availability, lab, previous publications and the conference where it was published.
> .. may one day pose similar risks due to the run-away prestige games scientists play.
It is happening in front of our eyes. From where should we start.
(1) Responsible AI, but the outcomes didn't challenge AI research enough to stop some of the practices. I would avoid commenting on what happened at Google AI recently, but I would say that most of the AI-based solutions that are deployed at large scale can be challenged.
(2) paperswithcode.com solved this issue. Back in the day, we would re-implement models that we see in the paper. When we email authors, we will always figure that an RA implemented the solution and s/he left.
(4) Rich is becoming richer. There is a phenomena in academia to blindly trust those who are in good labs. I have seen many papers that are great coming from these labs, but there are bad papers as well. we shouldn't just blindly trust ppl.
(5) Review process is broken. Review is a voluntarily work for professors, and imho it shouldn't. In industrial R&D, reviews are one of the most important things as part of the work. Some of our PRs takes days.
(6) All recommendation engines are black-boxes. I don't know what YT or FB recommend to a kid and whether this is aligned with some of our values. If s/he accidentally watch something bad, we noticed that their timeline is full by the same ideas. In our spare time, we tried to check if we can influence this behavior as a user. I figured that all of our recommendations engines are black-boxes. I don't know why we don't challenges. I hacked an App to fetch data, new, feed and run it through recommendation engine and started customizing what I can read.
(7) It's 2021, most of good papers are coming from industrial labs.
I agree that (4) is a problem but there is also some signal in brand name. There is such a flood of (mostly low quality) papers that it's impossible to look at their content to judge what's worth reading. If you want to be at the forefront of research you can't wait to see which one stand the test of time.
That's why I fall back on trusting brand name labs. They are "staking" their reputation on a paper. If a paper turns out to be absolute BS their reputation suffers. Even if all talent is equally distributed, this makes it more likely that papers from brand name researchers and institutions are carefully reviewed because they have something to lose. This isn't right, but what is the alternative?
(7) Because most resarch happens in industry labs. Many university academics have left, or at least have dual positions in university and industry.
(2) IMO paperswithcode didn't solve much. It's nice, but just publishing code doesn't fix any of the incentive problems. You still don't know how that code was generated. Most likely by tuning random knobs until something worked.
I agree with most of what you're saying, but 7 seems way off. Maybe it's field-dependent? I would say that in e.g. mathematics, physics and neuroscience, almost all good papers come from academic labs.
There are “industrial labs” for mathematics? (I would think that an implicit qualifier here is “in fields where there is actually an incentive for industry to do field-advancing research, to create + own the resulting IP.”)
I can't imagine two things making less tangible ("industrial") value than Mathematics and Neuroscience today. Physics is too big a field to make such an asinine comment on at least
Less tangible value than mathematics? The last few decades have been a golden age of linear and constraint programming research. The impact they've had on logistics alone is incalculable - it's changed how everyone does business from Ford and Fedex down to Uber drivers. Almost every mass manufacturer in the world uses them in one way or another to optimize their processes. I can't imagine an academic field that has had more impact on industry.
Most trivial and ubiqitous example is CAM which finds the path a CNC machine tool should take to translate a 3D model into a real part as efficiently as possible. At this point probably every physical product on the market involves CAM work somewhere in its production, either for the product or the machinery making it.
Thanks for answering, however I tried to search for "linear programming in CAM" and also skimmed wikipedia on CAM and linear programming and came up with nothing. Can you please point me to some resource on this?
My point is that the only counterexamples you could find to "industry labs > research labs" were in completely non-industrial scenarios like Mathematics and Neuroscience, two famously theoretical disciplines
Mathematics might be plenty large in theory-space, but real-world applicability is much more the domain of physics and thus I would expect many real-world applications to arise from it. So, to me, it's clear one cannot say Physics doesn't deliver on the physical from time to time
I had no idea the article was littered with links, as the colour difference is too small to see for me and my monitor. They look like the same colour to me. If Substack feels a different colour for links is too distracting or tacky, maybe they could consider a little icon denoting a link. But it's not very good web etiquette to hide links like that.
On the web, links are underlined by default, a clear, non-color way to visually convey they're links. It has become way too common for sites to remove that styling, maybe showing the underline when the cursor is hovered (useless on mobile devices without a cursor); this one doesn't even show the styling on :focus to make the visible on keyboard navigation.
The underline is an affordance, an indicator that it's special. The problem with removing it is there is no way to see it's a link without affirmatively taking an action over the text. Plus, on touch screens, there is no hovering.
While most screen reader users have some vision, screen reader software is mostly not affected by styling changes so they will still say "link" when there is one.
People with various kinds of low vision (or color blindness) who do not use a screen reader but use text enlargement, screen zooming, etc. would have a particularly hard time differentiating the links from the surrounding text.
> If you ask a gain-of-function proponent, they will say that by creating viruses that might emerge in nature, you get to understand zoonotic jumps from animals to humans better and possibly prevent them. Specifically, you get a head-start on developing vaccines for them. This possibility of curing future diseases might be true in some cases. But Pfizer’s COVID-19 vaccine was developed in a few hours back in January 2020.
I've been wondering about this for the last few months, I feel like this "better prepare for future viruses" argument is substantially weaker now — coronaviruses were a well-known group of viruses, there was already research directed at gain-of-function work for this family of viruses, and yet as far as I'm aware the mRNA vaccines that were developed derived no benefit from any of that research. So whether or not the virus came from a lab, why should we fund this kind of work since it seems not to be very useful?
There's more to GoF research than developing vaccines. Sometimes its trying to understand host virus interaction, virus evolution, zoonotic events. The list goes on. The majority of primary research is not at all interested in vaccine development.
For coronaviruses they're a huge family of viruses. Sometimes it's not transferable between species
The argument the post is presenting seems like it could be a straw man. Gain-of-function research would surely be useful to things other than vaccines, which might even be more important than vaccines? For example, figuring out in what conditions viruses jump hosts and therefore how to reduce the likelihood of new epidemics.
> The only consensus is no consensus. We may never know.
Well that's refreshing. Of course, the preceding paragraph gives a lot of strong evidence for a lab link.
> Baseless conspiracy theories will run wild, like the American government did it on purpose... or that the Chinese government released it on their own citizens. These aren’t true.
And here again we fall into the "things I don't like can't possibly be true" trap, which is just as ridiculous as the wild-eye'd absolutist conspiracy theory position. Consider: There were military games in Wuhan just before the outbreak.[1] Apparently intelligence knew about the coronavirus as early as November[2]. Iran's senior leadership was hit very hard by the virus[3].
Do these facts prove that the virus was intentionally released? Of course not. But we have to at least consider the possibility. If it was intentionally released, does that mean the US or China did it? Of course not. It could have been a rogue group within either of those two countries, another country entirely (trying to stir up tensions?), or a non-governmental group.
As the author noted previously: we may never know.
Science prestige games suck. But the lynchpin of this thesis is the assertion that gain-of-function research didn't (and won't) improve our ability to respond to epidemics with vaccines and policies. That point isn't substantially argued in the post, author just says that one COVID vaccine was developed in hours.
I think the author just has a complaint about the scientific process (and believe me, all scientists do…) and is trying to fit this topic to that complaint. I doubt that it’s helpful. If we think there aren’t enough safeguards about this kind of research, I think the solution is to restrict funding, not to assume that changing the research incentive structure would stop it.
One might accuse the author of playing a game of his own, where his writing is driven by a desire to get clicks rather than a desire to communicate factually...
IMHO research about human body in general and immune system in specific is more useful to improve our ability to respond to epidemics with vaccines than gain-of-function research.
The Moderna covid vaccine was developed in a few hours, because extensive work had already been done on the SARS virus. They could very easily tell by homology which part of the RNA sequence coded for the spike protein. I suppose theoretically, generating novel viruses could prepare us for similarly quick vaccine design in the absence of a natural predecessor, but that would be a huge stretch requiring us to invent vast numbers of novel viruses.
Difficult part is to isolate an unknown virus and sequebce its RNA and proteins. With covid we got lucky because Chinese scientists published that very fast. But it could easily take several months.
As a researcher I would gladly do this. Unfortunately the funding game doesn't support this at all.
I agree that papers are not good for knowledge transfer, but they are required if you want to keep a roof on your head.
Academic community seems to be unable do little else than complain about this. Even though at least in countries I'm familiar with (Finland and UK), academic community mostly runs the funding system as well, although some outside politics do shape it too.
I think what is needed is public pressure to academia and politicians to fix this mess. Unfortunately any criticism of science from the outside seems to be thrown into conspiracy theory category, but this is done mostly by non-academics.
It's not that bad personally, I can and do do demos and show code. Just have to do the chore of writing a paper, but otherwise in academia people are quite free to do what they want. I find it to be more a disservice to society that funds academia.
If COVID-19 did come from a (accidental) leak from gain of function research, presumably the pandemic itself counts as a demo. Which leads to the posted article's point---it may be better to not do any gain of function research on viruses that are already well-understood, regardless of how well the research is or isn't communicated.
Papers are good for communicating the big picture: what is being done, why certain choices were made, and how things are supposed to work. Writing them is useful, because it forces you to think about the big picture. It's common that people who have spent months or years working on a project suddenly understand some aspects of it better when they try to communicate it to other people.
Code is good for communicating details, but very few people actually read research code. Those who do usually just check a few details, because it's hard to justify spending weeks or months studying a single project. It doesn't help that the code was probably written before anyone understood what was the exact problem they were trying to solve and what would be a good way of solving it.
> It is immensely unfortunate this issue became politicized.
This comment is over a year too late. Almost immediately after COVID became the topic du jour the technocrats and media were aggressively shutting down anyone who noticed the circumstantial evidence that the coronavirus emerged in close proximity to a lab conducting sloppy coronavirus research. That, coupled with China's refusal to allow anyone outside the CCP apparatus to investigate should have, at least, given thinking people pause. But clearly, "thinking people" doesn't characterize the media or the technocrats and their speech police.
And we shouldn't forget the astonishingly sudden about-face by the media (almost all of whom work for companies who also sell their entertainment properties in China) who routinely called COVID the "China virus" or "Wuhan virus" one day and the next day castigated anyone who dared to do so as an irredeemable racist.
A lot of this “circumstantial evidence” was and is game-of-telephone nonsense. Many mix together two completely different institutions (the Wuhan Center for Disease Control and the Wuhan Institute of Virology) which are several kilometres apart, for example.
I’m not sure this is a valid counterargument. The intent is clear when someone says “it likely leaked from the lab in Wuhan”… I’m surprised how pedantic everything has gotten.
It's actually this exact attitude that kept it out of the media. I'm curious: what is your goal? Everyone can see the glaringly obvious circumstantial evidence that it was a lab leak. At this point, comments like your's just look ridiculous.
Oh, so vulnerable and fragile media decided to spin a ridiculous narrative because of stupid comments on the internet? Maybe the media should grow up and stop acting like a stereotypical 13 year old girl.
To be honest, the "evidence" I've seen so far boils down to stating that a virology lab was nearby the first outbreak so sure it must have come from that lab and couldn't have come from a wet market or elsewhere. In other words, evidence is confused with plausible storytelling, in incredibly common mistake nowadays.
What's mildly infuriating me about this is that the people who push the lab escape hypothesis seem to have political motivations, but the wet market origin hypothesis is much worse for China. Although they have officially prohibited many of the practices and animal abuse that make it easier for viruses to jump to humans from other hosts, these practices remain widespread. It's much easier to secure virology labs than tens of thousands of semi-legal and illegal wet markets.
The fascinating part is everyone did an about face on it overnight after the WSJ article.
We went from Facebook banning mention of the theory to everyone being certain it's right on the basis of "national security flack says someone who works at the lab was sick, during flu season".
> The WSJ author had brought us other CIA-approved gems about WMDs in Iraq once upon a time.
In case other people besides myself were confused, we're apparently talking about Michael Gordon, and this WSJ article [0], not this one [1]. Michael Gordon also wrote [2], which reported on the Bush administration's claims that Iraq was developing nuclear weapons.
I downvoted you, and I'll explain why. In my recollection, the very first "politicization" I heard of Covid's origins were politicians insisting that the virus was man-made. Politicians, mind you, who had been repeatedly clear that they were not particularly interested in figuring out the truth in what happened, but assigning blame.
So then I saw the unfortunate (and this is where I agree with part of your sentiment) response from media and the scientific establishment insisting that this wasn't possible, instead of stating the truth, that is "we just don't know". Note these are similar falsehoods that the scientific establishment took early on in the pandemic, saying "masks don't work for the public", instead of saying that "we're scared there won't be enough masks for hospitals, and we don't know how effective masks are by the general public."
Finally, this statement by you is downright laughable:
> And we shouldn't forget the astonishingly sudden about-face by the media (almost all of whom work for companies who also sell their entertainment properties in China) who routinely called COVID the "China virus" or "Wuhan virus"
I recall quite clearly one particular American president calling it "the China virus" and virtually every mainstream US media outlet calling him out for racism. I don't remember any member of the media establishment calling it the China virus.
The corporate media is good at memory holing things, but here's a compilation of them calling it the "Wuhan Coronavirus", "Chinese Coronavirus" and "China's Coronavirus":
- Link A: The clips are a compilation of COVID19 being referred to as the "Wuhan Coronavirus" and "China Coronavirus" in Q1 2020. I'm not sure what I was supposed to take away from it, is the implication that The US Media was hypercritical because people criticized the US president for repeatedly saying it xenophobically, months later?
- Link B: The clips are a compilation of health experts on MSNBC/CNN in January 2020, when the outbreak was confined to China, telling people they didn't need to worry yet.
I thought we'd emerge from this with an understanding of how quickly things move and operating under uncertainty a la Taleb, but in the current American context, it appears it turned into another reason to triple down on They'ing.
> The clips are a compilation of health experts in January 2020
Did you watch the video? Christiane Amanpour is not a health expert.
The media joined the outrage mob only when the wrong people started calling it such. That's more evidence of the early politicization driven by the same media.
The argument as described seems weak: "Combination of words $X said on Any News Outlet at any time mean its impossible for anyone to ever criticize $X being said by anyone else in any context at any time after. If it happens, Politicized Media."
I wonder if there's a bit more nuanced way to make this argument, I'm amenable to it, but...it reads more as American culture war than a discussion
> I wonder if there's a bit more nuanced way to make this argument, I'm amenable to it, but...it reads more as American culture war than a discussion
That's a fair critique, but I'm inclined to leave a more nuanced argument in someone else's hands. I'll just say I've long suspected massive media collusion and I found the media's almost overnight religious conversion on speech that identified the source of the virus to be jarring and suspiciously in sync.
There are dozens of egregious examples of the media adopting certain language or using certain terms in a sudden, widespread manner. A few examples you can find on Youtube include "power through", "beginning of the end", "walls closing in", etc.
I do think the media monoculture is partially to blame, but I don't that accounts for all of it.
I think it occurs someone criticizes a political opponent for using a word or phrase which has been commonplace until that time, though some may have considered it offensive.
Immediately, that usage is widely decried as horribly offensive, and even groups that routinely used it pile on to join the attack:
"When (my opponent) said X today, it shows they are horrible and intend terrible things, whereas my use of it yesterday/last year (or in the future!) was completely innocent."
The direction of the politicization will remain in dispute:
Did my opponent politicize it by using language known to be offensive, as a callout to their constituency?
Or did I politicize it by criticizing them, as a callout to my constituency?
Or do they politicize it by continuing to use it afterward?
It was first called the Wuhan Virus by mass media and researchers, even in China. That naming choice was fairly ordinary, similar to Spanish Flu, West Nile Virus, or Crimean–Congo fever. The "China virus" term became problematic once it was politicized in a criminally dumb way by the Trump government - back when they claimed it's not dangerous at all and totally under control in the US.
Insisting on a more neutral name was helpful for distinguishing reliable information from nonsense. But in my opinion the name Sars-Cov-2 was unhelpful in the long run because people just ended up calling it Covid or Coronavirus instead. Wuhan virus would have been a better naming choice, and it was called like that inside China at the beginning. The politicization of the health crisis (instead of showing solidarity first) was the big issue. Worse, it's still ongoing.
If totally neutral names are so important, maybe virologists should start thinking about naming schemes like meteorologists use, e.g. using first names.
>In my recollection, the very first "politicization" I heard of Covid's origins were politicians insisting that the virus was man-made.
I'd like to see some evidence that politicians insisted it was man-made instead of trusting your recollection. If the virus had lab origins in a lab specifically performing "gain of function" research, that's quite a coincidence that might even suggest those politicians were correct (although I would concede it may have certainly been irresponsible).
Additionally, I expect polemics and politicization from politicians. It is their nature. But I'm not supposed to get it from the media, from the CDC, from the technocrats, and from the "fact checkers". But their rhetoric was political and their actions to shut down and shout down wrongthink was wrong (and now can be seen as having specious justification).
> I don't remember any member of the media establishment calling it the China virus.
Another commenter already explained this, but these types of videos that are deliberately designed to obfuscate what actually happened are bullshit.
Yes, the media called it the "Wuhan coronavirus" early on before it was an actual pandemic because the virus primarily only existed in Wuhan at that point. Furthermore, the now well-known name of Covid-19 was not as prevalent then, and your average listener would understand "coronavirus" more than a new technical name at that point.
This is very different than, after the pandemic had already been widespread, specifically crossing out "Coronavirus" to call it "Chinese virus", as Trump did.
Look, there are plenty of things that the media at large deservedly gets criticized for, but trying to pretend calling it the Wuhan coronavirus early on in the pandemic is the same thing as specifically emphasizing it as "the Chinese virus" 6 months later is just lying.
> Look, there are plenty of things that the media at large deservedly gets criticized for, but trying to pretend calling it the Wuhan coronavirus early on in the pandemic is the same thing as specifically emphasizing it as "the Chinese virus" 6 months later is just lying.
Your argument explicitly depends on the context and it depends on the criticism. The context wasn't "6 months later". It was days later. And the criticism was that it is racist. That it is xenophobic. That it is intended to humiliate and threaten people of Chinese descent.
A much more plausible explanation would have been that Trump and his insistence on calling it the "China virus" was an attempt to deflect any political blame for his administration's policies to try and contain the virus. A more plausible explanation was that Trump was being unnecessarily coarse. Or unnecessarily provocative. But no, he was accused of racism and xenophobia.
But even before the phrasing of "China virus" was a thing, Trump was criticized for shutting down travel from China as being racist. In hindsight it turns out to have been a very shrewd, life-saving move. But instead of giving credit where credit's due and recognizing the wisdom of shutting down travel, suddenly the phrasing was racist and xenophobic. The accusations were the same, but the goalposts were moved. That the same people (the media) lobbing the accusations of racism and xenophobia were only days earlier using the same language is clear hypocrisy.
Since racism is wrongthink and wrongthink doesn't merit being discredited, it's easy to see why it's thrown around so casually. It requires no justification and is a cheap way to generate outrage. And that makes it a useful political tool.
We now know that many of these same people (the media along with their cohorts in government and the socials) shut down reasonable discussion regarding the origins of the virus while simultaneously using accusations of racism and xenophobia. And that's potentially scandalous and should be deeply embarrassing.
I also recall mass confusion about what constitutes “man-made”. That could mean anything from a natural-origin virus released accidentally from a lab that was legitimately studying it, to an artificial-origin virus deliberately released from a lab for malign purposes.
(As well as the even milder variant of “natural-origin virus that didn’t leak from a lab, but which humans encountered as a result of sample-gathering for research”)
It was also political because of the punitive measures sought by those pointing fingers to China— ranging from 10 trillion in reparations to war. Rather than saying hey, maybe this gain of function business was a mistake & we should just do it in the supercomputers like with nukes or something.
Personally, I don't really understand the focus on pinning down the exact reason this happened. Stop GoF research and shut down wet markets. Stop encroaching on wilderness as best we can while we're at it. Do all the things.
Many reforms will be aggressively resisted. Wet markets are unlikely to be shut down unless they can be plausibly fingered as the source of COVID-19 (and maybe not even then). GoF research has been heavily criticized for years, but it is still funded because there's a lobby in favour of it. People saying "hey, this could cause a pandemic" wasn't enough. There's another in favour of encroaching on wilderness.
> Do all the things.
We're more likely going to do none of the things. Again, unless one of them can be shown to be directly related to COVID, and even then, it's questionable.
Maybe after 2020, the other side of those arguments have a bit more weight for everybody. Were we to focus more on those things rather than war and reparation, maybe we can survive as a species. But yeah, probably not.
Is it realistic to expect them to comply with demands that they should stop GOF research / shut down wet markets without a credible threat of (at least) an economic retribution?
The US does this kind of research, too. And, it seems like it even funded the Wuhan research at least to some extent. Would the US need a threat of economic retribution to reconsider this research? Or to change a practice of their food supply suspected of causing significant harm? Writing that out, I guess maybe it would.
> who routinely called COVID the "China virus" or "Wuhan virus" one day and the next day castigated anyone who dared to do so as an irredeemable racist.
Probably because Patient Zero was located in Wuhan? What sort of cognitive dissonance am I missing here?
I used to work in one of the top fMRI labs. The comments about The Science Game are spot on. The vast majority of scientists, by the time they're midway through their postdocs, just want to do whatever research will get them a professorship and some grants. Actual knowledge generation is a distant second to keeping the ball bouncing by getting out papers and getting in grants.
Just want to add, it's generally not a conscious choice. Everyone wants to do something that is both a genius experiment and gets published in Nature. The Science Game gets propagated in how scientists conceive and plan their work. Everything must be done in the paradigm of publications and grants. You can't do anything if it doesn't have at least interim results within a year or two. It's an unfathomable career risk to try something absolutely brand new. There's a good chance it fails, at which point you have a 2 year blank spot in your CV and you can never get back in the top 10% of scientists where you need to be to get R01 funding and run your own lab.
One problem is the incentives are biased toward “ok” papers.
h-index should be replaced with a more competitive index (e.g., Max over citations). The publication hyper-competitiveness is due to publication spam, which boosts citation and publication count. However, twisting a knob and getting slightly different results should be viewed as a failed project.
> The publication hyper-competitiveness is due to publication spam, which boosts citation and publication count.
The publication hyper-competitiveness is a direct result of grant hyper-competitiveness. Funding rates at top US agencies are around 10-20%. Grants are how researchers fund their research, pay their students, pay themselves, and earn tenure, so there's a lot riding on getting them.
I see a lot of talk about the problems with publications on HN and possible fixes to the publication process (like requiring code for example, or as you suggest changing the important metrics), but publications really are just a proxy (and prerequisite) for the grant application process. You can change around the publication process all you want, but nothing will improve unless you address the issue of grants, because that's where the money is.
Right. Listing 10 papers with “ok” impact should be viewed as 10 “bad” papers and not 10 “good” papers.
It’s extremely unlikely that a 50 page CV is demonstrating real value throughout. It’s the academic equivalent of putting 20 Javascript frameworks on your Resume because you used them one time.
I agree that h-index isn't very indicative, but max citations seems like a poor alternative.
There is a relation between citations and certain aspects of research quality, but citation count can be easily gamed, and is affected by many factors other than the quality of the work itself.
Trying to use single measures to quantiy something as complex seems flawed to begin with.
Top N papers is already a metric used for professorship slots, so Max or other functions are me simplifying the situation.
I agree though that citations are proxies for goodness and are ultimately poor indications of work. For example, a paper closing the field would have no follow-up work, but a provocative paper would have plenty (regardless of quality).
I think a quantitative metric is necessary because many decisions eventually hit a bean-counter (e.g., admin type), who is not able to assess research quality in the time they have. But even with experts reviewing a set of candidates on research qualifications, the panel would be unable to objectively provide a full ranking of candidates.
A perhaps better way to do it would be to perform proper accounting on research output which stabilizes across a 10-year timeline or so. Each researcher is a “stock” and the stock-price reflects current and projected “revenue”. If someone publishes only papers that go nowhere, their stock tanks. At top institutions, tenure kind of does this (only one time).
> When a measure becomes a target, it ceases to be a good measure. - Goodhart's Law
> The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor. - Campbell's Law
You want to use metrics as a very broad breadth first search to help cull the search space and use trust systems as a depth first search. But once you have found signal through trust, you should completely ignore that first search and even look more into things that you had previously excluded.
Find researchers that you agree with, find the papers that they recommend, read those papers and check that you agree with them. If they recommend something from a no-name university, look at those first.
Unfortunately if this is not your field and you aren't able to determine quality, this becomes impossible. If it's important to you, you need to learn it. This is why I don't like non-technical managers. If the people who approve the grants do not understand the result, this is inevitable. It might work early on when trust still lingers, but as metrics take over the social systems always fall apart.
I think you're right. There's only a few people with the curiosity, tenacity, and obsessive patience to keep up their idealism and enjoyment momentum to deliver consistent intellectual quality for its own sake. You basically have to be on the spectrum or very, very diligent. Most people just want more funding. (ProTip: Full-time grant writers are key.)
> You basically have to be on the spectrum or very, very diligent.
...and be independently wealthy (or have wealthy benefactors). Science lives and dies by funding. We may be regressing to pre-20th century science, where most of the big discoveries were made by rich people who could afford to do science all day everyday.
Hmm, I do not see this at all. 40 years in neuroscience and genetics and it is far larger and more open community now that is also much more open to diversity than it was in 1978.
Hundreds of collaborators and none have supported their own research.
I am sympathetic to many of the ideas in this thread, but view many comments as too skeptical of the process and sociology of science and of career paths in science. The high entry barrier and generally low incomes compared to alternatives means an enrichment for individuals with genuine intellectual curiosity. Yes, publishing papers and getting grants involves some gaming, but that is also true of almost all forms of communication and interaction. Science is usually a game with a good cause.
Over the next 50 years we will get far beyond our still primitive methods of publishing key data and our analysis (aka papers). You can see the transition already in the form of web services, Jupyter notebooks, R/Shiny, Pluto, etc. Soon most core publications will become living, growing structures instead sad mausoleums. Data and code quality will become much more important when data and workflows and logic are highly visible and highly granular in a new post-FAIR+ science world.
It's really both. At least some of the marketing hype, maybe most of the hype, in the scientific community tends to coalesce around ideas that are obvious already, or that are on everyone's minds, or are appealing for very risk-averse reasons (something that allows one to churn out papers). The funding is risk averse along these same lines, in that it tends to follow slightly dated hype.
Do you think your thesis would be controversial within the Scientific Community, at least if discussions were to be had in public?
Does the Scientific community have enough self-awareness about this to do something about it, and are calls for 'creating new viruses' being actively made and addressed?
I actually think most working scientists agree about The Science Game. Almost everyone thinks of things this way in their darker moments, or talks among other scientists like that, although it is a bit of an "open secret." For gain-of-function research in particular, I don't imagine it will survive this news cycle. It has been banned for periods of time before (I believe last ban ended in 2017) and there's almost no way it doesn't get and likely stay banned, at least in the USA.
What path in life led you to write articles like this? If you are still a practicing scientist, do you fear any repercussions to speaking against the community like this?
Great article, I think an important one as well. The 'Science Game', really just a misalignment of incentives, seems to be at the core of other issues in science research - the replication problem in social sciences comes to mind (for me, as a non-academic). What do you see as possible solution(s) to this problem?
I'm a post doc virology researcher whose gain of function research is, ironically, delayed by covid.
I think you're spot on with the description of the Science Game. It's depressing, and I'm considering leaving academia because I can't stand the perverse incentive structures.
I think you're misguided with respect to why gain of function research is done. The idea that we're doing gain of function in order to generate more stuff to study is sort of right, but missing the point. It's not that we've run out of natural stuff to study. We have plenty of natural viruses to study. It's instead that gain of function is a great way to practically test some hypotheses. Imagine you have a theory that property X of a virus enables it to replicate in humans. You can remove property X from an existing virus, but it's even more convincing if you can predict what adding property X to a different virus will do.
Also I would argue that "gain of function research" is too broad a term to be useful for making policy. It's like discussing whether "nuclear physics" is dangerous. Building nuclear weapons is dangerous, making a particle accelerator isn't.
Virus and therapeutics researchers, for future history books:
Ron Fouchier, Netherlands
Yoshihiro Kawaoka, Japan
Alexander Kagansky, Russia (died 2020)
Bing Liu, USA (died 2020)
Frank Plummer, Canada (died 2019)
Gita Ramjee, South Africa (died 2020)
I've seen so many posts about or at least implicitly accepting the lab leak hypothesis in a very short time now, that it almost seems like some sort of concerted campaign to me. Would be interesting to know if there is an upvoting block doing this. Did I miss any new evidence coming out? Everything I saw so far seemed rather circumstantial.
This is a pretty good summary of the lab leak hypothesis: https://www.vanityfair.com/news/2021/06/the-lab-leak-theory-...
Evaluate it on its own merits. Personally I think it's incredibly compelling, but there are certainly many smart people who disagree.
I don’t really have an opinion one way or the other to offer on your concerted campaign comment or the origins of the covid 19 virus but my understanding is basically this…
People postulated
Trump politicized
Others kept researching and found important facts
There was internal fighting and dissension throughout various entities about whether or not to pursue these inquiries
Certain initial assessments may have been tainted by the biases of their authors
And we’re in the stage now where the inconsistencies, biases, and details are all coming out in this steady trickle and things are snowballing a bit.
The latest commit to that anonymously authored repo is over a year ago. I am a bit tired of these "we are an anonymous group pushing the truth" efforts; in the case of HCQ they are unmitigated bullshit.
The best sourced argument for lab leak is the Drastic project[1]. This is basically the work that's been getting the media attention for the last few weeks.
A good source for arguments holding the lab leak unlikely (there is of course no evidence strong enough to rule it out entirely) is episodes 760 and 762 of the podcast This Week in Virology[2]. Skeptics of authority would say that this is the community of virologists covering their own back. Others who basically believe in science [ETA: meaning basically trust scientists; this is a pretty complex concept to unpack] will find passionate and knowledgeable experts explaining why they find the lab leak hypothesis unlikely.
I grant you that this is well researched. However, readers should bear in mind that this resource is first and foremost a persuasive effort in favor of the lab leak hypothesis.
I suspect there are many people who consider it likely, felt silenced, and are now feeling vindicated.
While I obviously don't have hard evidence, the constant coverups make me consider the lab leak theory rather likely. The WHO inquiry seemed to conclude that they have no clue, but they were very adamant that it definitely wasn't a lab leak, with no supporting evidence. The whole thing reeked like an attempt to appease China to possibly get some more access.
Nope, not one bit. All pathogen research and engineering should be treated as a proliferation issue similar to nuclear weapons with international oversight. It's actually much, much worse because of independent biohacking and not needing rare enriched materials, which isn't a valid reason (chicken-little excuse) to not regulate it strongly globally.
No, if we have some evidence, it can inform outcomes and we can act.
This isn't a Judicial Trial, it's an investigation and pragmatic response.
If we're 85% sure it came from the lab, that's materially actionable.
But there's a 100% that China is obfuscating the situation on purpose and at very minimum they must be removed from the WHO. It's unthinkable that they could be a part of an organization and act directly against it at the same time.
They'll do as they please (in this and other things), without consequence, because there are simply not cost to their actions.
The issue is not Science, it's China. Given the existential consequences of COVID, the response to their intransigence should be existential, far beyond being removed from the WHO.
To your point, it was widely understood last year it could have been a lab leak but almost certainly not a bioweapon, and a lot of people a year later are turning the science of this into another culture war battle based on the idea that last year it was _censored_ that it could have been a lab leak.
I'm increasingly worried about a phenomena I see more and more often: a cohort 'rediscovering' an old topic and it being played as news. ex. Bitfinex/Tether was a scandal _years and years_ ago, and a lot of people rediscovered it and think its new because the legal proceedings from that initial outburst years ago finally wrapped up.
The theories seem to be: 1) Zoonotic 2) We have no idea, 3) Lab leak 4) Purposefully created and released bioweapon.
What seems to have happened is that there's a large group that have always backed theory 2, but theories 2, 3, and 4 were lumped together, and then all effectively suppressed, meaning all you saw being discussed was theory 1, despite the fact that it was always a minority view.
What happened fairly recently is that, for various reasons, theories 2 and 3 is no longer being effectively treated as synonymous with theory 4, and thus it's now "safe" to discuss theory 2 and 3. And now the large group of people who always backed theory 2 are seemingly everywhere.
(I'd caution against interpreting people supporting what I've labelled theory 2 as "implicitly accepting the lab leak hypothesis". As the linked article states, "...the only conclusion here is that of doubt. There is no direct evidence that COVID-19 escaped from the Wuhan Institute of Virology.")
Part of the confusion is that lab leak theory includes a zoonotic component. The virus started as a bat virus and then was modified in the lab. When Trump says "china made this virus" people assume he is saying it was made from scratch or made maliciously. Both of which are unlikely and Trump has a history of saying thing that are total bullshit like "my investigators in Hawaii are going to find proof of Obama's birth"
Trump's non-stop intentional effort to say outrageous and nonsensical things in order to inflame passions (for his own personal profit) and try to hide his own ignorance and lack of work ethic, make it important to have an honest public conversation. This can't be understated. It went so far that half a million US Americans died due a botched response that included people refusing to take basic safety precautions (and supported by their political leaders) just to "to own the libs"
It's still based on indicative evidence rather than definitive court-of-law evidence, but it's a good systematic attempt and makes me personally believe that the lab leak theory is more likely than not.
People dont call it the way it should be called: "gain of function" is basically biological weapon research.
You make a new disease and (supposedly) try to find a treatment for it. Yet somehow we dont hear about new vaccines or drugs found in those labs.
Since research on biological weapons is banned by conventions - bad actors try to overcome this by using non military labs.
Russia had Biopreparat who tested multiple viruses and bacteria and then made warheads to deliver them (long after it was banned). USA's Reston "accident" with monkeys also looks like reaserch on ebola. China makes "gain of function" in Wuhan, sometimes 300 meters from the wet market testrd as place of origin of the COVID19 virus.
What is gained here? Only military can gain something.
Those scientists are enemies od humanity - making biological weapons, but nobody will do anything about that. Because supposedly this research can lead to new drugs. Well... where are those new drugs or vaccines?
China, Russia and probably USA have those labs - a lot of money is spent, yet all real vaccines and drugs come from private labs. Not military.
You can read Ken Alibek's book on his work as one of the heads of Russian biological weapons program (hidden as normal research): the guy got infected with a militarized strain of tularemia - and instead of reporting it and going yo quarantaine, he just took some antibiotics and called in sick. The negligence of someone who knows about bio weapons was astounding.
You cannot rule out that same did not happen in China. In fact the press points to it: gain of funcion done in BSL2 labs.
Not to mention that USA reasearchers are on China's payroll, what looks like plain treason.
No, you can absolutely do gain of function research that is not dangerous. Gain of function merely means that you're adding some capability to a virus.
It doesn't mean it has to be a virus capable of infecting humans. It doesn't mean it has to make the virus more virulent in vivo.
It could be engineering bacteriophages to control resistant bacteria, for example.
"I ended up receiving my PhD in neuroscience at the University of Wisconsin-Madison, where I worked with Giulio Tononi on developing aspects of Integrated Information Theory. Later I was a postdoctoral researcher at Columbia University working with Rafael Yuste, as well as a visiting scholar at the Institute for Advanced Study in Princeton."
This article is a general critique of the reward system in academia, bolted on the lab leak hypothesis for clickbait. "The Science Game" is a thing and often discussed on HN, but I don't see any causative link here.
> All to say: scientists create dangerous synthetic viruses to achieve “high-impact” scientific output.
Does gain of function research yield more high impact papers than more benign types of virology research? The article seems to suggest so, but provides no citations indicating that. Were the scientists at the Wuhan Institute of Virology publishing their research to general acclaim up until now? Given how little we know about it, doesn't seem so. Are these scientists driven by the same invectives as western university research academics? They're probably state employees on stable contracts for a start, not the PhD students and itinerant PostDocs of the university system, who are the main players of the Science Game.
I don't know the answers to these questions, but the article would be more persuasive if it did.
> Does gain of function research yield more high impact papers than more benign types of virology research?
The article puts "high-impact" in quotes. The thesis as I understood it is not that gain-of-function research is higher impact, but that it's _a_ fertile knob capable of stamping out adequate papers. The externalities here should put it out of bounds of the Science Game. The author is basically saying, go find other knobs to play your games.
I'm skeptical of the article's point, but I think it does try to answer this question of yours:
> Does gain of function research yield more high impact papers than more benign types of virology research? The article seems to suggest so, but provides no citations indicating that.
The article's argument is that gain-of-function research moves the researcher from the set of existing viruses to the set of all the viruses we can create - a far, far larger set, and hence one that allows more publications. Or, to put it another way, after you've run out of viruses to find, and things to publish about them, creating more gives you many more papers to write.
There may be something to that argument, but I still agree with your general point that the article needs more evidence.
The actual impact of gain-of-function research papers isn't actually what is being critiqued here. That's putting the cart before the horse. What is being criticized is that the authors of such papers are pushing their research efforts into this area because they see it as their highest expected value avenue for generating the "impact" they need to secure funding, tenure, etc.
But if the gain-of-function research really is high-impact in the sense that is valuable to society, isn't it a good thing that incentives push scientists to work in this area? It would be an example of careerism "impact" lining up with actual impact; i.e., the system working as intended.
> if the gain-of-function research really is high-impact in the sense that is valuable to society
So far its net impact seems to be extremely negative (one pandemic, zero help with any treatments or preventive measures). So by your own argument, we should stop it immediately, right?
I'm not sure whether on the balance gain of function research is a good idea or not. The point of my comment was just that discussing the impact is highly relevant, and isn't "putting the cart before the horse". But I guess I should expand on it.
Regarding whether gain of function research is a good idea: Creating synthetic variants (to see if there is gain or loss in function) is very helpful in that it allows well-controlled experiments, which is necessary to establish causation. As I said elsewhere, I don't think the risk is appreciably different whether the lab's collection of dangerous viruses is all-natural or not. That is, if we ban gain of function research we should probably stop studying the unmodified variants also. There have been near-misses with accidental releases of SARS-CoV-1 and Guanarito [0]; no gain of function required. And the 1977 release of H1N1 came from vaccine development.
Regarding describing the net impact of gain of function research as one pandemic, zero help:
I don't know what would be different if gain of function research had been banned. The underlying methods have been used since at least 1999 on, among other things, MERS and SARS-2003 [0], so they have contributed to the general knowledge base that permitted extremely rapid development of the COVID-19 vaccine. This was a significant help, not zero help. On a technical and scientific level we were very well prepared for COVID-19. Its biology was well-understood almost immediately. Our failures were on the social and organizational levels.
Was gain of function research genuinely necessary to achieve that level of preparedness? To answer that one of us would have to sift through the body of work on COVID-19 vaccine development and trialed treatments to see if the people involved used information from mutagenesis or infectious clone technology experiments (the work that is being referred to as "gain of function"). I haven't done that, so I can't give a definitive reply on that point.
There is also the matter of side effects from a ban. Creating synthetic variants within the scope of naturally occurring traits seems acceptable given the benefits. But there's no guarantee what a given edit will do in advance. If you ban all genetic engineering that could lead to gain of function then that will severely hamper research, including development of vaccines and antivirals.
So overall I guess I favor a restriction on deliberately increasing pathogenicity, virulence, and transmissibility beyond levels that occur in similar natural viruses, but allowing the possibility that this could still happen unintentionally. I also think synthetic variants should be destroyed as soon as their purpose is complete, and facilities where these viruses are studied should have a test/trace/isolate plan in continuous operation.
Edit: We also have to keep in mind that COVID-19 wasn't necessarily leaked from a lab, and even it was, it may be a naturally occurring variant. Which would make much of this speculation pointless.
> Creating synthetic variants (to see if there is gain or loss in function) is very helpful in that it allows well-controlled experiments
Only if, as I responded to another post of yours elsewhere in this discussion, the experiments consist of infecting humans with different strains and seeing what happens to them. And even then there are a huge number of possible confounders.
> There have been near-misses with accidental releases of SARS-CoV-1 and Guanarito [0]; no gain of function required
Indeed. The fact that this problem has been around for a while doesn't make it any less of a problem. Plus, those releases had a much smaller impact because those previous viruses were not as well adapted to human infection when they first appeared--a feature that SARS-CoV-2 does not share (see further comments below).
> I don't know what would be different if gain of function research had been banned
If the lab escape theory is correct, we would not have had this pandemic because the lab would not have done the research in the first place. Seems like a pretty major negative consequence to me.
> On a technical and scientific level we were very well prepared for COVID-19. Its biology was well-understood almost immediately. Our failures were on the social and organizational levels.
The decision to do gain of function research in the first place, particularly in a lab in China where US officials could not hope to have any useful oversight of safety procedures, was also a social and organizational failure. (So was our failure to get a vaccine widely distributed even though, as you note, we understood the biology of the virus very quickly.) Doesn't make the pandemic any less devastating.
> they have contributed to the general knowledge base that permitted extremely rapid development of the COVID-19 vaccine
I don't know what knowledge base you're talking about. DNA sequencing technology, and generating mRNA from a given DNA sequence, which is how Pfizer and Moderna were able to produce a COVID-19 vaccine in a matter of hours, were around before any of the research you refer to was done, and that research added nothing new.
If you mean knowledge about the spike protein being important, that was known before any gain of function research was done, so I don't see how that research helps. Simple research into "how does this type of virus infect a human cell" would have been enough.
> I guess I favor a restriction on deliberately increasing pathogenicity, virulence, and transmissibility beyond levels that occur in similar natural viruses, but allowing the possibility that this could still happen unintentionally. I also think synthetic variants should be destroyed as soon as their purpose is complete, and facilities where these viruses are studied should have a test/trace/isolate plan in continuous operation.
While these are nice ideas, the problem is that they would have to be implemented with the same grossly failed institutions that got us into this mess.
To me this is similar to an argument I agree with against allowing capital punishment in our society: while I agree in principle that there can be cases where capital punishment is justified, the institutions in our society have shown that they are incapable of exercising the kind of discipline required to make sure that, when a person is sentenced to death, we know to a moral certainty that they are guilty of a crime that merits that punishment. Similarly, while in principle it might be the case that we could gain benefits from dangerous research with viruses, the institutions in our society have shown that they are incapable of exercising the kind of discipline required to do that safely.
> COVID-19 wasn't necessarily leaked from a lab, and even it was, it may be a naturally occurring variant
The fact that even the earliest samples obtained, in late 2019, were already highly infective to humans, is a huge indicator to me that t...
I think you may be approaching the discussion more from the standpoint of where to assign blame specifically regarding COVID-19, and I'm more interested in the benefits from infectious clone technology and infection experiments in general, which covers the last 50 years [0], not just the work in 2R01AI110964 that is currently under criticism. So I yield on all Wuhan-related points.
One thing that really confuses me though is why you say human testing is a requirement, when almost all work is done in cell & animal models. Usually human tests are restricted to treatments, and only those that have already gone through cell and animal testing to ensure as much safety as possible. I just want to emphasize this point because the thought of doing infectious clone testing on people is awful and the Common Rule is meant to prevent this kind of abuse [1].
Regarding "previous accidentally released viruses, which were believed to be naturally occurring variants, did not have this [highly infectious to humans] property." The ones that are notable enough to make it into reports kind of do. The 1977 H1N1 leak affected "20-70% of those under 20 years of age in school or military camp outbreaks in the first year" [2]. There were also several leaks of the 2002 SARS virus that could easily have gone the same way COVID-19 (SARS-CoV-2) did. China was the source of several of these leaks, so feel free to use this as evidence for a lax safety culture.
> I think you may be approaching the discussion more from the standpoint of where to assign blame specifically regarding COVID-19
No, I'm approaching it from the standpoint of how to not have a pandemic like this happen again.
> I'm more interested in the benefits from infectious clone technology and infection experiments in general
What are the benefits? What benefits have we gained specifically from this research, that we wouldn't have gained without it?
> why you say human testing is a requirement, when almost all work is done in cell & animal models
Because we can't reliably predict what a virus will do to humans unless humans have been infected with it. Cell and animal models are nice, but they aren't the same as actual human patients.
> Usually human tests are restricted to treatments
Exactly. And that's for good reasons. And my point is that testing treatments isn't the same as testing the viruses themselves. The Wuhan lab wasn't testing treatments, they were testing viruses. So how treatments are tested, and how confident we feel that they are safe based on cell and animal testing, is irrelevant to work with viruses themselves.
> and only those that have already gone through cell and animal testing to ensure as much safety as possible.
And even then we find effects in humans that we didn't expect. And that's for treatments, where we have more control over the variables than we do with viruses themselves.
> The ones that are notable enough to make it into reports kind of do.
"Kind of" isn't the same thing. I'm not talking about how infective they got in the first year. I'm talking about the way SARS-CoV-2 was in the very first human patients we have records of, back in late 2019. The strains isolated from that time were already highly infective. Initial strains of previous viruses were not; they increased in infectivity as they spread in humans and evolved to adapt. That's what we expect to happen for a virus that naturally jumps to humans from some other source. But when we see a virus that is already highly infective in the first human cases we know of, that's an indication that it did not naturally jump to humans, but was specifically engineered in a lab to be more infective to humans.
> There were also several leaks of the 2002 SARS virus that could easily have gone the same way COVID-19 (SARS-CoV-2) did.
> allowing the possibility that this could still happen unintentionally
Btw, this is also a key point: if we are going to do research that could unintentionally result in the release of a deadly pathogen, we need to agree as a society in advance on what countermeasures we will put in place if that happens. Leaders of society should not be having to "wing it" about these things after the release has occurred. For example, if the only way to stop a pandemic once the release reaches a certain stage is to stop all international travel, then everyone needs to agree in advance that if there is a release, and it reaches that stage (which in the case of COVID-19 would have been roughly mid-January 2020, when it was clear that infected people were making it from China to other countries), all international travel gets stopped until the outbreak is contained, no exceptions. (Imagine if that had been done in January 2020--we might have had no pandemic at all, and a lot less economic disruption overall to boot, since China, if you believe their numbers, had their outbreak under control by February 2020.) And if that gives a lot of people second thoughts about whether such research should be allowed at all, if that kind of drastic consequence has to be planned for, good.
Skimming the "Understanding the Risk of Bat Coronavirus Emergence" grant [0] cited in the posted article suggests a potentially useful role for gain of function research, although I'm not sure if that method was actually used.
The intended work was to (1) look for unknown viruses in bats and sequence the virus' genomes, (2) screen people who live and work nearby to see which viruses cross over to people, and (3) test the infectiousness of these viruses in cell culture and animal studies in a laboratory setting. The point of #3 would be to systematically compare the properties of viral strains with slight genetic variation to reveal which parts of the genome are responsible for which outcomes (symptoms, which species can be infected, etc.). Information from the lab comparisons would be used to help interpret which naturally occurring viruses observed in the screening work are potentially dangerous. Work like this _could_ help prevent a pandemic. (It would be bitterly ironic if it instead caused a pandemic.)
Synthetic variation of the viral genome ("gain of function", but also loss of function) allows direct experimentation to be done in the lab (#3). Direct experimentation is very useful because it allows systematic comparison to establish cause & effect; which genes do what, and how they interact. If a large supply of naturally occurring variants are already available though this can also be achieved, with less certainty, by correlation. But if humanity is going to be proactive about studying viruses, I don't see much reason to avoid gain of function research. Collecting and studying wild viruses means that dangerous new strains will be present in labs regardless, so either way stringent safety controls, monitoring, and containment plans are absolutely necessary.
> Direct experimentation is very useful because it allows systematic comparison to establish cause & effect
Only if "experimentation" includes "infecting humans with various strains and seeing what happens". Is that what you're advocating?
> Collecting and studying wild viruses means that dangerous new strains will be present in labs regardless
Dangerous new strains that happened to evolve naturally is one thing.
Dangerous new strains that are purposely being created by experimenters for the express purpose of finding ones that are more infective is another.
> so either way stringent safety controls, monitoring, and containment plans are absolutely necessary.
This would appear to be an argument for not doing this type of research in the Wuhan lab, which evidently did not have sufficient controls in place. Moreover, I see no indication that any assessment of such controls at the lab, and confirmation that they were sufficient, was a prerequisite for funding this research. That seems insane.
Agree with this critique completely. Erik should have picked a much better context in neuroscience to make his points, not in virology. The trademark addition to the science game is offensive. Don’t think Wittgenstein added TMs to his new terms or phrases much.
It's a tangent on the example given, but the bigger question is: Does the origin of SARS-CoV-2 really matter at this point? I think it's moot and the lab speculation trope comes across as an agenda-driven conspiracy theory.
We can reduce our risks by having strict international conventions on pathogen labs and engineering, eliminating high-density meat agriculture, and reducing overuse of antibiotics/antimycotics/antivirals. Pragmatically, human pathogens should be treated with the same care as nuclear weapons as a matter of risk-management.
> Does the origin of SARS-CoV-2 really matter at this point?
I hear this question so often I'm beginning to get upset.
Yes, it absolutely 100% matters for the same reason it matters that we learn about anything. Why would we ever want to restrict our knowledge and be willfully ignorant? At this point I just assume the only reason is political. And when has that ever been a good thing for humanity or history?
For all we know China is purposefully manufacturing viruses as weapons (in fact there is already some evidence of this happening), and you think it doesn't matter?
Honestly none of the anti-China propaganda ever really seemed to do anything for me, but nothing has made me more leery of China than the wall of "we don't really need to know where covid came from" comments I see on the internet. I don't know or care if the person I am replying to is a shill, but I have never seen such a strong front defending ignorance before in my life.
JFC everything gets flagged now instead of anyone having a discussion. I'm directly responding to part of the comment which I believe to already be political, so I'm just trying to continue the conversation that OP started. Maybe if most comments suggesting we should explore this issue weren't so quickly flagged on HN it would make me more receptive to the idea this isn't an organized political front.
Eh. I would like to believe it, but pragmatist in me knows it is not true. I will add that willingly ignoring this facet is doing a disservice to pandemic handling. It simply cannot exist without political will and capital.
Political Concerns are inseparable from those things.
Judicial Concerns matter as well, and we don't just ignore the manslaughter of 10's of millions of people because of populist concerns over xenophobia.
If this were XYZ Corp. USA, and they killed 1 million Americans due to their lab accident - all of you (edit: 'all of us') would be screaming for life-sentences for the executives and a Trillion in damages for the victims, their families, and probably for a 'total overhaul' of relevant structures. And that wouldn't be political.
> In virology, there are only so many dials—only so many natural viruses. .... The big excitement is in finding a new virus
How is that supposed to jibe with articles like this in which, unless I badly misread things, the authors found FORTY FIVE THOUSAND new viruses? https://www.pnas.org/content/118/23/e2023202118
Your study seems to be looking at remnants of viruses “left behind” in our collective genome by analyzing datasets. The viruses are hypothetical in nature, and might not be complete (they remark in the abstract that some have nearly complete genomes).
Finding them would be worthy of a paper, some of them might be complete enough to be able to produce further insights, but at the end of the day there’s only so much you can do when you don’t have a live virus to “tweak the dials” so to speak. 45k is certainly a lot of potentially interesting viruses to study, sure, but it’s a finite supply and the quality of genomes will likely further reduce this number.
And that’s kind of where the OP comes back into play. Sure, these new viruses were found, but how much can they contribute to the “Science Game”? Especially compared to a live virus you can tweak and play with to your heart’s content to make whatever you want.
With respect, I think you need to re-skim that paper.
> Your study seems to be looking at remnants of viruses “left behind” in our collective genome by analyzing datasets.
You appear to be confusing the 'genome' and 'metagenome'. The genome does have plenty of viral remnants[0] that by and large are incomplete fragments.
The 'metagenome' in this case is taken from the Human Microbiome Project[1], which took samples of the microbiome from various regions of the bodies of various humans[2], and then sequenced basically everything in there that they could.
> might not be complete (they remark in the abstract that some have nearly complete genomes).
The only thing about completeness in the abstract I'm looking at is "with historically high per-genome completeness".
Later in the paper they write "A total of 14,034 contigs (31.2%) were estimated to be high-quality (90 to 100% complete)" which I'd call more than 'some'!
> but at the end of the day there’s only so much you can do when you don’t have a live virus to “tweak the dials” so to speak
Many viruses are not currently able to be cultivated. This doesn't mean that they aren't important, or that they can't be studied.
For a different example, consider the anelloviruses. From [3]:
"""
Anelloviruses are small, single stranded circular DNA viruses. They are extremely diverse and have not been associated with any disease so far. Strikingly, these small entities infect most probably the complete human population, and there are no convincing examples demonstrating viral clearance from infected individuals. The main transmission could be via fecal-oral or airway route, as infections occur at an early age. However, due to the lack of an appropriate culture system, the virus–host interactions remain enigmatic. Anelloviruses are obviously mysterious viruses, and their impact on human life is not yet known, but, with no evidence of a disease association, a potential beneficial effect on human health should also be investigated.
"""
The way I read this, you are almost certainly infected with anelloviruses, I am almost certainly infected with anelloviruses, we don't know how they're transmitted, we don't know what cells they target, and in fact we don't know very much about what it's doing in there at all.
> a finite supply
Well sure, and there's only so many hundreds of millions of years before the sun devours the earth.
Hoel writes:
"""
In virology, there are only so many dials—only so many natural viruses. And each is a source of competition, as famous labs make claims to various viruses to study and monopolize them by beating others to publication.
The big excitement is in finding a new virus, mapping the genome, figuring out its function and transmissibility, comparing to other viruses, etc."""
How many virology labs are out there? The American Society of Virology has about 2500 members[4] in the US, Canada, and Mexico. Multiply by 10 to bring in the rest of the world (surely an overestimate, if anything) gives you 25000, which is still less than the number of brand-new viruses found in this one study!
And that's just from a handful of samples focusing on one organism (albeit one of particular interest). Wiki lists 96 families of virus[5], some of which have dozens of subpages. Viruses are everywhere you look, and infect every kind of life on earth including each other[6]
I'm also not convinced that labs can "monopolize them by beating others to publication". Many viruses are worked by many labs. As just a quick example, I searched biorxiv for 'herpes' (it's a virus!) and of the first few papers that looked like virology I found authors affiliated with Cambridge (Departments of Pathology, Veterinary Medicine, and Medicine, as well as the Institute for Medical Research) [7], the European Molecular Biolo...
170 comments
[ 3.6 ms ] story [ 219 ms ] threadAfter which over 150 million people got sick and over 3 million people died. It seems possible that the regulatory rigidity responsible for the long roll out is a greater mortal threat than the virus itself. It's hard to imagine plausible scenarios in which a complete failure to regulate could have been more destructive. We have met the enemy and he is us. Apparently the most important technology that we lack to prevent a repeat isn't biotech, but the means to regulate the regulators.
Edit: especially taking into account one of the main tasks of the regulators was to decide _which_ vaccine to choose. If I remember correctly, there were hundreds of vaccines in development. In hindsight we know which ones were the right choice, but that's the result of rigoros, regulated, time-consuming testing.
Phase III trials for Pfizer took ~16 weeks. It has been 26 weeks since the US granted an EUA based on that trial. In the 26 weeks since then, ~14% of the world has been vaccinated.
And that's after we spent those 16 weeks of the trial building out manufacturing capacity.
We did skip a lot of steps on this vaccine. When we started mass administration of it, we had no idea whether it would prevent or reduce transmission or just symptomatic infection. We are still figuring out how long we expect immunity to last, whether we need boosters every six months or never.
Maybe once we have the technological capacity to manufacture and distribute 8 billion vaccine doses ~instantly it will make sense to skip the "does this even work?" step, but as long as that remains the bottleneck I have a hard time believing it.
Once this is possible¹, we can ask governments to keep some extra capacity unused to respond to emergencies. What is different from any previous tech, where extra capacity would be useless nearly every time.
1 - AKA, when the patents expire. I vote for greatly rewarding the people that developed them and taking the patents away, but I'm sure that won't happen, so I hope we don't see any larger pandemics for 15 years.
I think we have come a long way and pushed the boundaries substantially.
Just like ML research, we just skip the p-values.
To be fair to there are a lot of smells in DL papers, usually you can tell whether an approach is worth your time by looking at code availability, lab, previous publications and the conference where it was published.
It is happening in front of our eyes. From where should we start.
(1) Responsible AI, but the outcomes didn't challenge AI research enough to stop some of the practices. I would avoid commenting on what happened at Google AI recently, but I would say that most of the AI-based solutions that are deployed at large scale can be challenged.
(2) paperswithcode.com solved this issue. Back in the day, we would re-implement models that we see in the paper. When we email authors, we will always figure that an RA implemented the solution and s/he left.
(4) Rich is becoming richer. There is a phenomena in academia to blindly trust those who are in good labs. I have seen many papers that are great coming from these labs, but there are bad papers as well. we shouldn't just blindly trust ppl.
(5) Review process is broken. Review is a voluntarily work for professors, and imho it shouldn't. In industrial R&D, reviews are one of the most important things as part of the work. Some of our PRs takes days.
(6) All recommendation engines are black-boxes. I don't know what YT or FB recommend to a kid and whether this is aligned with some of our values. If s/he accidentally watch something bad, we noticed that their timeline is full by the same ideas. In our spare time, we tried to check if we can influence this behavior as a user. I figured that all of our recommendations engines are black-boxes. I don't know why we don't challenges. I hacked an App to fetch data, new, feed and run it through recommendation engine and started customizing what I can read.
(7) It's 2021, most of good papers are coming from industrial labs.
That's why I fall back on trusting brand name labs. They are "staking" their reputation on a paper. If a paper turns out to be absolute BS their reputation suffers. Even if all talent is equally distributed, this makes it more likely that papers from brand name researchers and institutions are carefully reviewed because they have something to lose. This isn't right, but what is the alternative?
(7) Because most resarch happens in industry labs. Many university academics have left, or at least have dual positions in university and industry.
(2) IMO paperswithcode didn't solve much. It's nice, but just publishing code doesn't fix any of the incentive problems. You still don't know how that code was generated. Most likely by tuning random knobs until something worked.
I can't imagine two things making less tangible ("industrial") value than Mathematics and Neuroscience today. Physics is too big a field to make such an asinine comment on at least
(I also find it amusing that you categorize physics as "too big to make such an asinine comment about", but mathematics not)
My point is that the only counterexamples you could find to "industry labs > research labs" were in completely non-industrial scenarios like Mathematics and Neuroscience, two famously theoretical disciplines
Mathematics might be plenty large in theory-space, but real-world applicability is much more the domain of physics and thus I would expect many real-world applications to arise from it. So, to me, it's clear one cannot say Physics doesn't deliver on the physical from time to time
[1] https://www.natureindex.com/news-blog/google-scholar-reveals...
[2] https://www.natureindex.com/news-blog/google-scholar-reveals...
People with various kinds of low vision (or color blindness) who do not use a screen reader but use text enlargement, screen zooming, etc. would have a particularly hard time differentiating the links from the surrounding text.
Who's responsible for the styling, the author or Substack itself?
I've been wondering about this for the last few months, I feel like this "better prepare for future viruses" argument is substantially weaker now — coronaviruses were a well-known group of viruses, there was already research directed at gain-of-function work for this family of viruses, and yet as far as I'm aware the mRNA vaccines that were developed derived no benefit from any of that research. So whether or not the virus came from a lab, why should we fund this kind of work since it seems not to be very useful?
For coronaviruses they're a huge family of viruses. Sometimes it's not transferable between species
Well that's refreshing. Of course, the preceding paragraph gives a lot of strong evidence for a lab link.
> Baseless conspiracy theories will run wild, like the American government did it on purpose... or that the Chinese government released it on their own citizens. These aren’t true.
And here again we fall into the "things I don't like can't possibly be true" trap, which is just as ridiculous as the wild-eye'd absolutist conspiracy theory position. Consider: There were military games in Wuhan just before the outbreak.[1] Apparently intelligence knew about the coronavirus as early as November[2]. Iran's senior leadership was hit very hard by the virus[3].
Do these facts prove that the virus was intentionally released? Of course not. But we have to at least consider the possibility. If it was intentionally released, does that mean the US or China did it? Of course not. It could have been a rogue group within either of those two countries, another country entirely (trying to stir up tensions?), or a non-governmental group.
As the author noted previously: we may never know.
[1] - https://en.wikipedia.org/wiki/2019_Military_World_Games
[2] - https://abcnews.go.com/Politics/intelligence-report-warned-c...
[3] - https://www.wsj.com/articles/as-irans-leaders-contract-coron...
Anyone near the field able to give a perspective?
Difficult part is to isolate an unknown virus and sequebce its RNA and proteins. With covid we got lucky because Chinese scientists published that very fast. But it could easily take several months.
show me the code and or a demo.
I agree that papers are not good for knowledge transfer, but they are required if you want to keep a roof on your head.
Academic community seems to be unable do little else than complain about this. Even though at least in countries I'm familiar with (Finland and UK), academic community mostly runs the funding system as well, although some outside politics do shape it too.
I think what is needed is public pressure to academia and politicians to fix this mess. Unfortunately any criticism of science from the outside seems to be thrown into conspiracy theory category, but this is done mostly by non-academics.
Code is good for communicating details, but very few people actually read research code. Those who do usually just check a few details, because it's hard to justify spending weeks or months studying a single project. It doesn't help that the code was probably written before anyone understood what was the exact problem they were trying to solve and what would be a good way of solving it.
This comment is over a year too late. Almost immediately after COVID became the topic du jour the technocrats and media were aggressively shutting down anyone who noticed the circumstantial evidence that the coronavirus emerged in close proximity to a lab conducting sloppy coronavirus research. That, coupled with China's refusal to allow anyone outside the CCP apparatus to investigate should have, at least, given thinking people pause. But clearly, "thinking people" doesn't characterize the media or the technocrats and their speech police.
And we shouldn't forget the astonishingly sudden about-face by the media (almost all of whom work for companies who also sell their entertainment properties in China) who routinely called COVID the "China virus" or "Wuhan virus" one day and the next day castigated anyone who dared to do so as an irredeemable racist.
It was political from the beginning.
(Edit: https://www.youtube.com/watch?v=ab-r0capbzk is a great look at said game of telephone.)
What's mildly infuriating me about this is that the people who push the lab escape hypothesis seem to have political motivations, but the wet market origin hypothesis is much worse for China. Although they have officially prohibited many of the practices and animal abuse that make it easier for viruses to jump to humans from other hosts, these practices remain widespread. It's much easier to secure virology labs than tens of thousands of semi-legal and illegal wet markets.
We went from Facebook banning mention of the theory to everyone being certain it's right on the basis of "national security flack says someone who works at the lab was sick, during flu season".
The WSJ author had brought us other CIA-approved gems about WMDs in Iraq once upon a time.
You might be onto something with garden-variety liberals feeling 'free' to try and blame China now that Trump's out of sight, though.
In case other people besides myself were confused, we're apparently talking about Michael Gordon, and this WSJ article [0], not this one [1]. Michael Gordon also wrote [2], which reported on the Bush administration's claims that Iraq was developing nuclear weapons.
[0] https://www.wsj.com/articles/intelligence-on-sick-staff-at-w... [1] https://www.wsj.com/articles/wuhan-lab-leak-question-chinese... [2] https://www.nytimes.com/2002/09/08/world/threats-responses-i...
So then I saw the unfortunate (and this is where I agree with part of your sentiment) response from media and the scientific establishment insisting that this wasn't possible, instead of stating the truth, that is "we just don't know". Note these are similar falsehoods that the scientific establishment took early on in the pandemic, saying "masks don't work for the public", instead of saying that "we're scared there won't be enough masks for hospitals, and we don't know how effective masks are by the general public."
Finally, this statement by you is downright laughable:
> And we shouldn't forget the astonishingly sudden about-face by the media (almost all of whom work for companies who also sell their entertainment properties in China) who routinely called COVID the "China virus" or "Wuhan virus"
I recall quite clearly one particular American president calling it "the China virus" and virtually every mainstream US media outlet calling him out for racism. I don't remember any member of the media establishment calling it the China virus.
https://www.youtube.com/watch?v=5eZtCq1aj2g
Bonus complication of when "the flu is worse" was the narrative:
https://www.youtube.com/watch?v=wVDPVBZF2Xg
- Link B: The clips are a compilation of health experts on MSNBC/CNN in January 2020, when the outbreak was confined to China, telling people they didn't need to worry yet.
I thought we'd emerge from this with an understanding of how quickly things move and operating under uncertainty a la Taleb, but in the current American context, it appears it turned into another reason to triple down on They'ing.
Did you watch the video? Christiane Amanpour is not a health expert.
The media joined the outrage mob only when the wrong people started calling it such. That's more evidence of the early politicization driven by the same media.
I wonder if there's a bit more nuanced way to make this argument, I'm amenable to it, but...it reads more as American culture war than a discussion
That's a fair critique, but I'm inclined to leave a more nuanced argument in someone else's hands. I'll just say I've long suspected massive media collusion and I found the media's almost overnight religious conversion on speech that identified the source of the virus to be jarring and suspiciously in sync.
There are dozens of egregious examples of the media adopting certain language or using certain terms in a sudden, widespread manner. A few examples you can find on Youtube include "power through", "beginning of the end", "walls closing in", etc.
I do think the media monoculture is partially to blame, but I don't that accounts for all of it.
Immediately, that usage is widely decried as horribly offensive, and even groups that routinely used it pile on to join the attack:
The direction of the politicization will remain in dispute:Insisting on a more neutral name was helpful for distinguishing reliable information from nonsense. But in my opinion the name Sars-Cov-2 was unhelpful in the long run because people just ended up calling it Covid or Coronavirus instead. Wuhan virus would have been a better naming choice, and it was called like that inside China at the beginning. The politicization of the health crisis (instead of showing solidarity first) was the big issue. Worse, it's still ongoing.
If totally neutral names are so important, maybe virologists should start thinking about naming schemes like meteorologists use, e.g. using first names.
I'd like to see some evidence that politicians insisted it was man-made instead of trusting your recollection. If the virus had lab origins in a lab specifically performing "gain of function" research, that's quite a coincidence that might even suggest those politicians were correct (although I would concede it may have certainly been irresponsible).
Additionally, I expect polemics and politicization from politicians. It is their nature. But I'm not supposed to get it from the media, from the CDC, from the technocrats, and from the "fact checkers". But their rhetoric was political and their actions to shut down and shout down wrongthink was wrong (and now can be seen as having specious justification).
> I don't remember any member of the media establishment calling it the China virus.
https://www.youtube.com/watch?v=5eZtCq1aj2g
Another commenter already explained this, but these types of videos that are deliberately designed to obfuscate what actually happened are bullshit.
Yes, the media called it the "Wuhan coronavirus" early on before it was an actual pandemic because the virus primarily only existed in Wuhan at that point. Furthermore, the now well-known name of Covid-19 was not as prevalent then, and your average listener would understand "coronavirus" more than a new technical name at that point.
This is very different than, after the pandemic had already been widespread, specifically crossing out "Coronavirus" to call it "Chinese virus", as Trump did.
Look, there are plenty of things that the media at large deservedly gets criticized for, but trying to pretend calling it the Wuhan coronavirus early on in the pandemic is the same thing as specifically emphasizing it as "the Chinese virus" 6 months later is just lying.
Your argument explicitly depends on the context and it depends on the criticism. The context wasn't "6 months later". It was days later. And the criticism was that it is racist. That it is xenophobic. That it is intended to humiliate and threaten people of Chinese descent.
A much more plausible explanation would have been that Trump and his insistence on calling it the "China virus" was an attempt to deflect any political blame for his administration's policies to try and contain the virus. A more plausible explanation was that Trump was being unnecessarily coarse. Or unnecessarily provocative. But no, he was accused of racism and xenophobia.
But even before the phrasing of "China virus" was a thing, Trump was criticized for shutting down travel from China as being racist. In hindsight it turns out to have been a very shrewd, life-saving move. But instead of giving credit where credit's due and recognizing the wisdom of shutting down travel, suddenly the phrasing was racist and xenophobic. The accusations were the same, but the goalposts were moved. That the same people (the media) lobbing the accusations of racism and xenophobia were only days earlier using the same language is clear hypocrisy.
Since racism is wrongthink and wrongthink doesn't merit being discredited, it's easy to see why it's thrown around so casually. It requires no justification and is a cheap way to generate outrage. And that makes it a useful political tool.
We now know that many of these same people (the media along with their cohorts in government and the socials) shut down reasonable discussion regarding the origins of the virus while simultaneously using accusations of racism and xenophobia. And that's potentially scandalous and should be deeply embarrassing.
(As well as the even milder variant of “natural-origin virus that didn’t leak from a lab, but which humans encountered as a result of sample-gathering for research”)
Personally, I don't really understand the focus on pinning down the exact reason this happened. Stop GoF research and shut down wet markets. Stop encroaching on wilderness as best we can while we're at it. Do all the things.
> Do all the things.
We're more likely going to do none of the things. Again, unless one of them can be shown to be directly related to COVID, and even then, it's questionable.
Probably because Patient Zero was located in Wuhan? What sort of cognitive dissonance am I missing here?
h-index should be replaced with a more competitive index (e.g., Max over citations). The publication hyper-competitiveness is due to publication spam, which boosts citation and publication count. However, twisting a knob and getting slightly different results should be viewed as a failed project.
The publication hyper-competitiveness is a direct result of grant hyper-competitiveness. Funding rates at top US agencies are around 10-20%. Grants are how researchers fund their research, pay their students, pay themselves, and earn tenure, so there's a lot riding on getting them.
I see a lot of talk about the problems with publications on HN and possible fixes to the publication process (like requiring code for example, or as you suggest changing the important metrics), but publications really are just a proxy (and prerequisite) for the grant application process. You can change around the publication process all you want, but nothing will improve unless you address the issue of grants, because that's where the money is.
It’s extremely unlikely that a 50 page CV is demonstrating real value throughout. It’s the academic equivalent of putting 20 Javascript frameworks on your Resume because you used them one time.
There is a relation between citations and certain aspects of research quality, but citation count can be easily gamed, and is affected by many factors other than the quality of the work itself.
Trying to use single measures to quantiy something as complex seems flawed to begin with.
I agree though that citations are proxies for goodness and are ultimately poor indications of work. For example, a paper closing the field would have no follow-up work, but a provocative paper would have plenty (regardless of quality).
I think a quantitative metric is necessary because many decisions eventually hit a bean-counter (e.g., admin type), who is not able to assess research quality in the time they have. But even with experts reviewing a set of candidates on research qualifications, the panel would be unable to objectively provide a full ranking of candidates.
A perhaps better way to do it would be to perform proper accounting on research output which stabilizes across a 10-year timeline or so. Each researcher is a “stock” and the stock-price reflects current and projected “revenue”. If someone publishes only papers that go nowhere, their stock tanks. At top institutions, tenure kind of does this (only one time).
> When a measure becomes a target, it ceases to be a good measure. - Goodhart's Law
> The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor. - Campbell's Law
You want to use metrics as a very broad breadth first search to help cull the search space and use trust systems as a depth first search. But once you have found signal through trust, you should completely ignore that first search and even look more into things that you had previously excluded.
Find researchers that you agree with, find the papers that they recommend, read those papers and check that you agree with them. If they recommend something from a no-name university, look at those first.
Unfortunately if this is not your field and you aren't able to determine quality, this becomes impossible. If it's important to you, you need to learn it. This is why I don't like non-technical managers. If the people who approve the grants do not understand the result, this is inevitable. It might work early on when trust still lingers, but as metrics take over the social systems always fall apart.
...and be independently wealthy (or have wealthy benefactors). Science lives and dies by funding. We may be regressing to pre-20th century science, where most of the big discoveries were made by rich people who could afford to do science all day everyday.
Hundreds of collaborators and none have supported their own research.
Over the next 50 years we will get far beyond our still primitive methods of publishing key data and our analysis (aka papers). You can see the transition already in the form of web services, Jupyter notebooks, R/Shiny, Pluto, etc. Soon most core publications will become living, growing structures instead sad mausoleums. Data and code quality will become much more important when data and workflows and logic are highly visible and highly granular in a new post-FAIR+ science world.
Does the Scientific community have enough self-awareness about this to do something about it, and are calls for 'creating new viruses' being actively made and addressed?
Note sure if that's your choice or Substack's, but either way, it could stand improvement.
(Unfortunately, styling annoyances stand between content and topical response and criticism.)
https://news.ycombinator.com/item?id=27412599
I think you're spot on with the description of the Science Game. It's depressing, and I'm considering leaving academia because I can't stand the perverse incentive structures.
I think you're misguided with respect to why gain of function research is done. The idea that we're doing gain of function in order to generate more stuff to study is sort of right, but missing the point. It's not that we've run out of natural stuff to study. We have plenty of natural viruses to study. It's instead that gain of function is a great way to practically test some hypotheses. Imagine you have a theory that property X of a virus enables it to replicate in humans. You can remove property X from an existing virus, but it's even more convincing if you can predict what adding property X to a different virus will do.
Also I would argue that "gain of function research" is too broad a term to be useful for making policy. It's like discussing whether "nuclear physics" is dangerous. Building nuclear weapons is dangerous, making a particle accelerator isn't.
People postulated
Trump politicized
Others kept researching and found important facts
There was internal fighting and dissension throughout various entities about whether or not to pursue these inquiries
Certain initial assessments may have been tainted by the biases of their authors
And we’re in the stage now where the inconsistencies, biases, and details are all coming out in this steady trickle and things are snowballing a bit.
The best sourced argument for lab leak is the Drastic project[1]. This is basically the work that's been getting the media attention for the last few weeks.
A good source for arguments holding the lab leak unlikely (there is of course no evidence strong enough to rule it out entirely) is episodes 760 and 762 of the podcast This Week in Virology[2]. Skeptics of authority would say that this is the community of virologists covering their own back. Others who basically believe in science [ETA: meaning basically trust scientists; this is a pretty complex concept to unpack] will find passionate and knowledgeable experts explaining why they find the lab leak hypothesis unlikely.
[1]: https://drasticresearch.org/
[2]: https://www.microbe.tv/twiv/
While I obviously don't have hard evidence, the constant coverups make me consider the lab leak theory rather likely. The WHO inquiry seemed to conclude that they have no clue, but they were very adamant that it definitely wasn't a lab leak, with no supporting evidence. The whole thing reeked like an attempt to appease China to possibly get some more access.
Closing the stable door after the horse has bolted.
This isn't a Judicial Trial, it's an investigation and pragmatic response.
If we're 85% sure it came from the lab, that's materially actionable.
But there's a 100% that China is obfuscating the situation on purpose and at very minimum they must be removed from the WHO. It's unthinkable that they could be a part of an organization and act directly against it at the same time.
They'll do as they please (in this and other things), without consequence, because there are simply not cost to their actions.
The issue is not Science, it's China. Given the existential consequences of COVID, the response to their intransigence should be existential, far beyond being removed from the WHO.
I'm increasingly worried about a phenomena I see more and more often: a cohort 'rediscovering' an old topic and it being played as news. ex. Bitfinex/Tether was a scandal _years and years_ ago, and a lot of people rediscovered it and think its new because the legal proceedings from that initial outburst years ago finally wrapped up.
What seems to have happened is that there's a large group that have always backed theory 2, but theories 2, 3, and 4 were lumped together, and then all effectively suppressed, meaning all you saw being discussed was theory 1, despite the fact that it was always a minority view.
What happened fairly recently is that, for various reasons, theories 2 and 3 is no longer being effectively treated as synonymous with theory 4, and thus it's now "safe" to discuss theory 2 and 3. And now the large group of people who always backed theory 2 are seemingly everywhere.
(I'd caution against interpreting people supporting what I've labelled theory 2 as "implicitly accepting the lab leak hypothesis". As the linked article states, "...the only conclusion here is that of doubt. There is no direct evidence that COVID-19 escaped from the Wuhan Institute of Virology.")
'We don't know' can't be a theory (by definition). Theories try to explain and know.
It's still based on indicative evidence rather than definitive court-of-law evidence, but it's a good systematic attempt and makes me personally believe that the lab leak theory is more likely than not.
You make a new disease and (supposedly) try to find a treatment for it. Yet somehow we dont hear about new vaccines or drugs found in those labs.
Since research on biological weapons is banned by conventions - bad actors try to overcome this by using non military labs. Russia had Biopreparat who tested multiple viruses and bacteria and then made warheads to deliver them (long after it was banned). USA's Reston "accident" with monkeys also looks like reaserch on ebola. China makes "gain of function" in Wuhan, sometimes 300 meters from the wet market testrd as place of origin of the COVID19 virus.
What is gained here? Only military can gain something.
Those scientists are enemies od humanity - making biological weapons, but nobody will do anything about that. Because supposedly this research can lead to new drugs. Well... where are those new drugs or vaccines? China, Russia and probably USA have those labs - a lot of money is spent, yet all real vaccines and drugs come from private labs. Not military.
You can read Ken Alibek's book on his work as one of the heads of Russian biological weapons program (hidden as normal research): the guy got infected with a militarized strain of tularemia - and instead of reporting it and going yo quarantaine, he just took some antibiotics and called in sick. The negligence of someone who knows about bio weapons was astounding. You cannot rule out that same did not happen in China. In fact the press points to it: gain of funcion done in BSL2 labs.
Not to mention that USA reasearchers are on China's payroll, what looks like plain treason.
It doesn't mean it has to be a virus capable of infecting humans. It doesn't mean it has to make the virus more virulent in vivo.
It could be engineering bacteriophages to control resistant bacteria, for example.
https://www.erikphoel.com/about.html
"I ended up receiving my PhD in neuroscience at the University of Wisconsin-Madison, where I worked with Giulio Tononi on developing aspects of Integrated Information Theory. Later I was a postdoctoral researcher at Columbia University working with Rafael Yuste, as well as a visiting scholar at the Institute for Advanced Study in Princeton."
> All to say: scientists create dangerous synthetic viruses to achieve “high-impact” scientific output.
Does gain of function research yield more high impact papers than more benign types of virology research? The article seems to suggest so, but provides no citations indicating that. Were the scientists at the Wuhan Institute of Virology publishing their research to general acclaim up until now? Given how little we know about it, doesn't seem so. Are these scientists driven by the same invectives as western university research academics? They're probably state employees on stable contracts for a start, not the PhD students and itinerant PostDocs of the university system, who are the main players of the Science Game.
I don't know the answers to these questions, but the article would be more persuasive if it did.
The article puts "high-impact" in quotes. The thesis as I understood it is not that gain-of-function research is higher impact, but that it's _a_ fertile knob capable of stamping out adequate papers. The externalities here should put it out of bounds of the Science Game. The author is basically saying, go find other knobs to play your games.
> Does gain of function research yield more high impact papers than more benign types of virology research? The article seems to suggest so, but provides no citations indicating that.
The article's argument is that gain-of-function research moves the researcher from the set of existing viruses to the set of all the viruses we can create - a far, far larger set, and hence one that allows more publications. Or, to put it another way, after you've run out of viruses to find, and things to publish about them, creating more gives you many more papers to write.
There may be something to that argument, but I still agree with your general point that the article needs more evidence.
So far its net impact seems to be extremely negative (one pandemic, zero help with any treatments or preventive measures). So by your own argument, we should stop it immediately, right?
Regarding whether gain of function research is a good idea: Creating synthetic variants (to see if there is gain or loss in function) is very helpful in that it allows well-controlled experiments, which is necessary to establish causation. As I said elsewhere, I don't think the risk is appreciably different whether the lab's collection of dangerous viruses is all-natural or not. That is, if we ban gain of function research we should probably stop studying the unmodified variants also. There have been near-misses with accidental releases of SARS-CoV-1 and Guanarito [0]; no gain of function required. And the 1977 release of H1N1 came from vaccine development.
Regarding describing the net impact of gain of function research as one pandemic, zero help:
I don't know what would be different if gain of function research had been banned. The underlying methods have been used since at least 1999 on, among other things, MERS and SARS-2003 [0], so they have contributed to the general knowledge base that permitted extremely rapid development of the COVID-19 vaccine. This was a significant help, not zero help. On a technical and scientific level we were very well prepared for COVID-19. Its biology was well-understood almost immediately. Our failures were on the social and organizational levels.
Was gain of function research genuinely necessary to achieve that level of preparedness? To answer that one of us would have to sift through the body of work on COVID-19 vaccine development and trialed treatments to see if the people involved used information from mutagenesis or infectious clone technology experiments (the work that is being referred to as "gain of function"). I haven't done that, so I can't give a definitive reply on that point.
There is also the matter of side effects from a ban. Creating synthetic variants within the scope of naturally occurring traits seems acceptable given the benefits. But there's no guarantee what a given edit will do in advance. If you ban all genetic engineering that could lead to gain of function then that will severely hamper research, including development of vaccines and antivirals.
So overall I guess I favor a restriction on deliberately increasing pathogenicity, virulence, and transmissibility beyond levels that occur in similar natural viruses, but allowing the possibility that this could still happen unintentionally. I also think synthetic variants should be destroyed as soon as their purpose is complete, and facilities where these viruses are studied should have a test/trace/isolate plan in continuous operation.
Edit: We also have to keep in mind that COVID-19 wasn't necessarily leaked from a lab, and even it was, it may be a naturally occurring variant. Which would make much of this speculation pointless.
[0] https://yurideigin.medium.com/lab-made-cov2-genealogy-throug...
Only if, as I responded to another post of yours elsewhere in this discussion, the experiments consist of infecting humans with different strains and seeing what happens to them. And even then there are a huge number of possible confounders.
> There have been near-misses with accidental releases of SARS-CoV-1 and Guanarito [0]; no gain of function required
Indeed. The fact that this problem has been around for a while doesn't make it any less of a problem. Plus, those releases had a much smaller impact because those previous viruses were not as well adapted to human infection when they first appeared--a feature that SARS-CoV-2 does not share (see further comments below).
> I don't know what would be different if gain of function research had been banned
If the lab escape theory is correct, we would not have had this pandemic because the lab would not have done the research in the first place. Seems like a pretty major negative consequence to me.
> On a technical and scientific level we were very well prepared for COVID-19. Its biology was well-understood almost immediately. Our failures were on the social and organizational levels.
The decision to do gain of function research in the first place, particularly in a lab in China where US officials could not hope to have any useful oversight of safety procedures, was also a social and organizational failure. (So was our failure to get a vaccine widely distributed even though, as you note, we understood the biology of the virus very quickly.) Doesn't make the pandemic any less devastating.
> they have contributed to the general knowledge base that permitted extremely rapid development of the COVID-19 vaccine
I don't know what knowledge base you're talking about. DNA sequencing technology, and generating mRNA from a given DNA sequence, which is how Pfizer and Moderna were able to produce a COVID-19 vaccine in a matter of hours, were around before any of the research you refer to was done, and that research added nothing new.
If you mean knowledge about the spike protein being important, that was known before any gain of function research was done, so I don't see how that research helps. Simple research into "how does this type of virus infect a human cell" would have been enough.
> I guess I favor a restriction on deliberately increasing pathogenicity, virulence, and transmissibility beyond levels that occur in similar natural viruses, but allowing the possibility that this could still happen unintentionally. I also think synthetic variants should be destroyed as soon as their purpose is complete, and facilities where these viruses are studied should have a test/trace/isolate plan in continuous operation.
While these are nice ideas, the problem is that they would have to be implemented with the same grossly failed institutions that got us into this mess.
To me this is similar to an argument I agree with against allowing capital punishment in our society: while I agree in principle that there can be cases where capital punishment is justified, the institutions in our society have shown that they are incapable of exercising the kind of discipline required to make sure that, when a person is sentenced to death, we know to a moral certainty that they are guilty of a crime that merits that punishment. Similarly, while in principle it might be the case that we could gain benefits from dangerous research with viruses, the institutions in our society have shown that they are incapable of exercising the kind of discipline required to do that safely.
> COVID-19 wasn't necessarily leaked from a lab, and even it was, it may be a naturally occurring variant
The fact that even the earliest samples obtained, in late 2019, were already highly infective to humans, is a huge indicator to me that t...
One thing that really confuses me though is why you say human testing is a requirement, when almost all work is done in cell & animal models. Usually human tests are restricted to treatments, and only those that have already gone through cell and animal testing to ensure as much safety as possible. I just want to emphasize this point because the thought of doing infectious clone testing on people is awful and the Common Rule is meant to prevent this kind of abuse [1].
Regarding "previous accidentally released viruses, which were believed to be naturally occurring variants, did not have this [highly infectious to humans] property." The ones that are notable enough to make it into reports kind of do. The 1977 H1N1 leak affected "20-70% of those under 20 years of age in school or military camp outbreaks in the first year" [2]. There were also several leaks of the 2002 SARS virus that could easily have gone the same way COVID-19 (SARS-CoV-2) did. China was the source of several of these leaks, so feel free to use this as evidence for a lax safety culture.
[0] https://www.virology.ws/2009/02/12/infectious-dna-clones/
[1] https://www.hhs.gov/ohrp/regulations-and-policy/regulations/...
[2] https://armscontrolcenter.org/wp-content/uploads/2016/02/Esc...
No, I'm approaching it from the standpoint of how to not have a pandemic like this happen again.
> I'm more interested in the benefits from infectious clone technology and infection experiments in general
What are the benefits? What benefits have we gained specifically from this research, that we wouldn't have gained without it?
> why you say human testing is a requirement, when almost all work is done in cell & animal models
Because we can't reliably predict what a virus will do to humans unless humans have been infected with it. Cell and animal models are nice, but they aren't the same as actual human patients.
> Usually human tests are restricted to treatments
Exactly. And that's for good reasons. And my point is that testing treatments isn't the same as testing the viruses themselves. The Wuhan lab wasn't testing treatments, they were testing viruses. So how treatments are tested, and how confident we feel that they are safe based on cell and animal testing, is irrelevant to work with viruses themselves.
> and only those that have already gone through cell and animal testing to ensure as much safety as possible.
And even then we find effects in humans that we didn't expect. And that's for treatments, where we have more control over the variables than we do with viruses themselves.
> The ones that are notable enough to make it into reports kind of do.
"Kind of" isn't the same thing. I'm not talking about how infective they got in the first year. I'm talking about the way SARS-CoV-2 was in the very first human patients we have records of, back in late 2019. The strains isolated from that time were already highly infective. Initial strains of previous viruses were not; they increased in infectivity as they spread in humans and evolved to adapt. That's what we expect to happen for a virus that naturally jumps to humans from some other source. But when we see a virus that is already highly infective in the first human cases we know of, that's an indication that it did not naturally jump to humans, but was specifically engineered in a lab to be more infective to humans.
> There were also several leaks of the 2002 SARS virus that could easily have gone the same way COVID-19 (SARS-CoV-2) did.
How do we know they could have?
Btw, this is also a key point: if we are going to do research that could unintentionally result in the release of a deadly pathogen, we need to agree as a society in advance on what countermeasures we will put in place if that happens. Leaders of society should not be having to "wing it" about these things after the release has occurred. For example, if the only way to stop a pandemic once the release reaches a certain stage is to stop all international travel, then everyone needs to agree in advance that if there is a release, and it reaches that stage (which in the case of COVID-19 would have been roughly mid-January 2020, when it was clear that infected people were making it from China to other countries), all international travel gets stopped until the outbreak is contained, no exceptions. (Imagine if that had been done in January 2020--we might have had no pandemic at all, and a lot less economic disruption overall to boot, since China, if you believe their numbers, had their outbreak under control by February 2020.) And if that gives a lot of people second thoughts about whether such research should be allowed at all, if that kind of drastic consequence has to be planned for, good.
The intended work was to (1) look for unknown viruses in bats and sequence the virus' genomes, (2) screen people who live and work nearby to see which viruses cross over to people, and (3) test the infectiousness of these viruses in cell culture and animal studies in a laboratory setting. The point of #3 would be to systematically compare the properties of viral strains with slight genetic variation to reveal which parts of the genome are responsible for which outcomes (symptoms, which species can be infected, etc.). Information from the lab comparisons would be used to help interpret which naturally occurring viruses observed in the screening work are potentially dangerous. Work like this _could_ help prevent a pandemic. (It would be bitterly ironic if it instead caused a pandemic.)
Synthetic variation of the viral genome ("gain of function", but also loss of function) allows direct experimentation to be done in the lab (#3). Direct experimentation is very useful because it allows systematic comparison to establish cause & effect; which genes do what, and how they interact. If a large supply of naturally occurring variants are already available though this can also be achieved, with less certainty, by correlation. But if humanity is going to be proactive about studying viruses, I don't see much reason to avoid gain of function research. Collecting and studying wild viruses means that dangerous new strains will be present in labs regardless, so either way stringent safety controls, monitoring, and containment plans are absolutely necessary.
[0] https://reporter.nih.gov/search/q4dXFDKEsU-IkTAYgowOKw/proje...
Only if "experimentation" includes "infecting humans with various strains and seeing what happens". Is that what you're advocating?
> Collecting and studying wild viruses means that dangerous new strains will be present in labs regardless
Dangerous new strains that happened to evolve naturally is one thing.
Dangerous new strains that are purposely being created by experimenters for the express purpose of finding ones that are more infective is another.
> so either way stringent safety controls, monitoring, and containment plans are absolutely necessary.
This would appear to be an argument for not doing this type of research in the Wuhan lab, which evidently did not have sufficient controls in place. Moreover, I see no indication that any assessment of such controls at the lab, and confirmation that they were sufficient, was a prerequisite for funding this research. That seems insane.
We can reduce our risks by having strict international conventions on pathogen labs and engineering, eliminating high-density meat agriculture, and reducing overuse of antibiotics/antimycotics/antivirals. Pragmatically, human pathogens should be treated with the same care as nuclear weapons as a matter of risk-management.
I hear this question so often I'm beginning to get upset.
Yes, it absolutely 100% matters for the same reason it matters that we learn about anything. Why would we ever want to restrict our knowledge and be willfully ignorant? At this point I just assume the only reason is political. And when has that ever been a good thing for humanity or history?
For all we know China is purposefully manufacturing viruses as weapons (in fact there is already some evidence of this happening), and you think it doesn't matter?
Honestly none of the anti-China propaganda ever really seemed to do anything for me, but nothing has made me more leery of China than the wall of "we don't really need to know where covid came from" comments I see on the internet. I don't know or care if the person I am replying to is a shill, but I have never seen such a strong front defending ignorance before in my life.
JFC everything gets flagged now instead of anyone having a discussion. I'm directly responding to part of the comment which I believe to already be political, so I'm just trying to continue the conversation that OP started. Maybe if most comments suggesting we should explore this issue weren't so quickly flagged on HN it would make me more receptive to the idea this isn't an organized political front.
And it does not make sense until you recognize potential political fallout from it.
How would it affect your opinion if covid19 gain of function research was funded by US?
How would it affect your opinion if it there is an actual proof of China developing this virus?
Both have their ramifications and they are not even mutually exclusive.
Judicial Concerns matter as well, and we don't just ignore the manslaughter of 10's of millions of people because of populist concerns over xenophobia.
If this were XYZ Corp. USA, and they killed 1 million Americans due to their lab accident - all of you (edit: 'all of us') would be screaming for life-sentences for the executives and a Trillion in damages for the victims, their families, and probably for a 'total overhaul' of relevant structures. And that wouldn't be political.
Matter for what?
The origin was extremely important as long as all the good people agreed that said origin had nothing to do with the Wuhan lab.
It's curious that the origin is not important now that said good people are saying that the Wuhan lab might be the origin.
It's almost like the importance of the origin depends on what the origin is.
Also I sacrificed two years of my life. I think I deserve an answer.
How is that supposed to jibe with articles like this in which, unless I badly misread things, the authors found FORTY FIVE THOUSAND new viruses? https://www.pnas.org/content/118/23/e2023202118
Finding them would be worthy of a paper, some of them might be complete enough to be able to produce further insights, but at the end of the day there’s only so much you can do when you don’t have a live virus to “tweak the dials” so to speak. 45k is certainly a lot of potentially interesting viruses to study, sure, but it’s a finite supply and the quality of genomes will likely further reduce this number.
And that’s kind of where the OP comes back into play. Sure, these new viruses were found, but how much can they contribute to the “Science Game”? Especially compared to a live virus you can tweak and play with to your heart’s content to make whatever you want.
> Your study seems to be looking at remnants of viruses “left behind” in our collective genome by analyzing datasets.
You appear to be confusing the 'genome' and 'metagenome'. The genome does have plenty of viral remnants[0] that by and large are incomplete fragments.
The 'metagenome' in this case is taken from the Human Microbiome Project[1], which took samples of the microbiome from various regions of the bodies of various humans[2], and then sequenced basically everything in there that they could.
> might not be complete (they remark in the abstract that some have nearly complete genomes).
The only thing about completeness in the abstract I'm looking at is "with historically high per-genome completeness".
Later in the paper they write "A total of 14,034 contigs (31.2%) were estimated to be high-quality (90 to 100% complete)" which I'd call more than 'some'!
> but at the end of the day there’s only so much you can do when you don’t have a live virus to “tweak the dials” so to speak
Many viruses are not currently able to be cultivated. This doesn't mean that they aren't important, or that they can't be studied.
For a different example, consider the anelloviruses. From [3]: """ Anelloviruses are small, single stranded circular DNA viruses. They are extremely diverse and have not been associated with any disease so far. Strikingly, these small entities infect most probably the complete human population, and there are no convincing examples demonstrating viral clearance from infected individuals. The main transmission could be via fecal-oral or airway route, as infections occur at an early age. However, due to the lack of an appropriate culture system, the virus–host interactions remain enigmatic. Anelloviruses are obviously mysterious viruses, and their impact on human life is not yet known, but, with no evidence of a disease association, a potential beneficial effect on human health should also be investigated. """
The way I read this, you are almost certainly infected with anelloviruses, I am almost certainly infected with anelloviruses, we don't know how they're transmitted, we don't know what cells they target, and in fact we don't know very much about what it's doing in there at all.
> a finite supply
Well sure, and there's only so many hundreds of millions of years before the sun devours the earth.
Hoel writes: """ In virology, there are only so many dials—only so many natural viruses. And each is a source of competition, as famous labs make claims to various viruses to study and monopolize them by beating others to publication. The big excitement is in finding a new virus, mapping the genome, figuring out its function and transmissibility, comparing to other viruses, etc."""
How many virology labs are out there? The American Society of Virology has about 2500 members[4] in the US, Canada, and Mexico. Multiply by 10 to bring in the rest of the world (surely an overestimate, if anything) gives you 25000, which is still less than the number of brand-new viruses found in this one study!
And that's just from a handful of samples focusing on one organism (albeit one of particular interest). Wiki lists 96 families of virus[5], some of which have dozens of subpages. Viruses are everywhere you look, and infect every kind of life on earth including each other[6]
I'm also not convinced that labs can "monopolize them by beating others to publication". Many viruses are worked by many labs. As just a quick example, I searched biorxiv for 'herpes' (it's a virus!) and of the first few papers that looked like virology I found authors affiliated with Cambridge (Departments of Pathology, Veterinary Medicine, and Medicine, as well as the Institute for Medical Research) [7], the European Molecular Biolo...