Bird Flu and Biomedicine, An Essay on What We Might Expect

Yves here. KLG turns to what should be getting more serious attentions than it is: bird flu and what we might do now, having not made serious interventions early. His intro:

As H5N1 spreads the scientific literature will ramp up, so we need to know what to expect. This post considers a preprint on the putative origins of the current H5N1 variant of concern. While the overall hypothesis is not out of bounds, the authors cannot be considered objective, disinterested observers. This has been common in the COVID-19 literature (now up to 426,491 entries in PubMed), with less than optimum results. So with H5N1 we need to read well and with close attention to the “priors” of those who contribute to the current literature on H5N1. This is not an easy task with so many publishers and purveyors in action in the current environment, but it must be done. Whatever we do, the mistakes made with COVID-19 must not be repeated with a spreading H5N1 avian flu.

By KLG, who has held research and academic positions in three US medical schools since 1995 and is currently Professor of Biochemistry and Associate Dean. He has performed and directed research on protein structure, function, and evolution; cell adhesion and motility; the mechanism of viral fusion proteins; and assembly of the vertebrate heart. He has served on national review panels of both public and private funding agencies, and his research and that of his students has been funded by the American Heart Association, American Cancer Society, and National Institutes of Health

Are we facing a new pandemic, less than five years after the beginning of the current one?  Maybe.  The H5N1 “bird flu” is causing much consternation among the medical establishment and naturally, given the disputed origin and trajectory of COVID-19, the origins of Highly Pathogenic Avian Influenza (HPAI) H5N1 are already being discussed and disputed by those who want to get ahead of the game. Unfortunately, this is the nature of the current literature of Biomedicine [1], as we have often discussed in this series for nearly the past two years.  Recent updates include these from the FDA, CDC, and WHO.

Several days ago I was forwarded a copy of a preprint with the title Proximal Origin of Epidemic Highly Pathogenic Avian Influenza H5N1 Clade and Spread by Migratory Waterfowl.  This has been an interesting topic with ramifications that characterize the current scientific literature.  As with virtually all preprints and peer-reviewed papers in the online databases, this paper looks quite “good.”  A preprint server produces a professional product.  This paper investigates:

(T)he possible laboratory origins of…HPAI H5N1…currently affecting various animal species and causing sporadic human infections…(and concludes that)…the proximal origins of HPAI H5N1 may be the USDA Southeast Poultry Research Laboratory (SEPRL) in Athens, Georgia, and the Erasmus Medical Center in Rotterdam…as the result of serial passage (of the virus) in mallard ducks at SEPRL.

The ostensible purpose of this paper is to pose a question and suggest further research:

It is (sic) possible that HPAI H%N1 clade evolved not in nature, but as a result of serial passage of other Gain-of Function (GOF) research in a laboratory?  We hope the following investigative report will serve as a starting point for further investigation by specialists in the fields of virology, molecular biology, and avian flu epidemiology.

There can be no serious objection to this on the surface.  The cult of the expert has led us astray regarding COVID-19 for more than four years, without much of a resolution in sight.  Their figures are well presented, but the authors are not convincing in their selective review of the literature and conclusion that this variant of H5N1 if the likely product of serial passage of the virus (or intentional mutagenesis of this variant).  Nor is it obvious how one variant can have two origins, one in Rotterdam and one in Georgia.

Still, it is clear that laboratory accidents do cause infections.  During my early apprenticeship in the laboratory, I read of a centrifuge service technician who contracted Rocky Mountain Spotted Tick Fever and died.  Our regular service contract technician would not go near our centrifuges for routine maintenance until they were sterilized.  Although we did not use any infectious agents, he was correct in his precaution.  This paper does raise questions that should be answered whether it is ever published or not after peer review. [2]  Similar questions regarding SARS-CoV-2 have not been answered.  And therein lies much of the backstory if this very preliminary contribution to the current H5N1 literature.

But first a primer on reading the scientific literature and the use and abuse of preprints.  At the beginning of my life as a scientist, a “preprint” was a paper in manuscript form that was in preparation or had already been submitted to a journal but not yet published.  It was circulated confidentially among collaborators and colleagues informally so that they could be aware of where the field was going.  Somewhat later preprints became the route to publication in mathematics and physics, as manuscripts were made available to the interested readers on arXiv (1991).  A paper could remain in preprint form for years before appearing officially “in print.”

More recently preprint servers have become common in the biological and biomedical sciences, with bioRxiv (2013) and medRxiv (2019) as successors to arXiv.  Both of these are hosted and moderated by Cold Spring Harbor Laboratory Press.  CSHL Press (distributed by Oxford University Press) is a division of Cold Spring Harbor Laboratory on Long Island.  With the lamentable exception of its detour into eugenics in from 1910 to 1939, CSHL has been one of the most productive biological laboratories in the world, with the institutional disinterestedness essential to the production of reliable science.  This was considered here previously in an analysis of the work of Nancy Cartwright and colleagues.

So now we come to the preprint under discussion, which is hosted by, a division of MDPI, whose open-access article processing charges (as of January 2023) are listed here.  The disclaimer, which is broadly applicable to all preprints, at the top of the pdf but not original link reads:

Disclaimer/Publisher’s Note: The statements, opinions, and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions, or products referred to in the content.

This is the first thing the general reader must keep in mind when reading the scientific literature, which is increasingly first publicized from preprints.  The second thing is to examine the interests and motives of the authors, and that includes all of the biomedical literature, preprints and published alike.

What about the present authors, who are affiliated with the McCullough Foundation of Dallas, Texas?  Nicolas Hulscher is a Fellow of the McCullough Foundation who produced this video (3:16) on the McCullough Foundation and “will be concerned with will provide insightful analysis across the four core pillars of the Foundation: Investigative Scholarship, Education, Justice, and Public Policy”.  John Leake studied with the late Sir Roger Scruton [3] and wrote The Courage to Face COVID-19 with Peter A. McCullough, who is the namesake of the McCullough Foundation and a well-known COVID contrarian (not that there is anything wrong with that in principle).  From the mission statement of the McCullough Foundation:

McCullough Foundation is dedicated to the proposition that a government of the people, by the people, for the people, can only exist if the people are educated about health, disease, and public health policies.  As James Madison, author of the U.S. Constitution remarked: “Knowledge will forever govern ignorance, and a people who mean to be their own governors must arm themselves with the power knowledge gives.”

As we learned during the COVID-19 pandemic, our deep-seated fear of deadly infectious disease makes us very susceptible to manipulation by public health officials (often in league with undeclared commercial interests) who invoke emergency power with the purported objective of protecting us. Thus, our fear of infectious disease can be exploited by unscrupulous public officials in the same way our fear of foreign invaders has been used by dictators and tyrants throughout history. Again, to quote James Madison: “The means of defense against foreign danger have been always the instruments of tyranny at home.”

This is standard libertarian discourse, such as it is.  This also finds its echo on the liberal side of the divide, for example in a significant selection of the catalog of Chelsea Green, one of my favorite publishers.  This did not arise ex nihilo.  Dr. McCullough is an advocate of what can only be called libertarian medicine and has recently been the Chief Scientific Officer of The Wellness Company, which among other things offered “free medical care” to the residents of East Palestine, Ohio after the horrific Norfolk Southern derailment in February 2023.

The question for us is “Where does libertarian medicine come from?”  Libertarian medicine should be a category mistake.  The practice of medicine has always been as much art as science, but in our infatuation with the idea of progress (here, for example) we have forgotten this.  Medicine has become (very) big business during my lifetime, with care of the people somewhere under the bottom line.  Consequently, the clinical judgment of primary care physicians, whose life’s purpose is to care for their patients as people and not conditions defined by algorithms, has been devalued.  The backlash has been severe among those who are the victims of our misguided healthcare system.  This includes those who deliver healthcare and those who need the help of their doctors and nurses.

From the beginning of COVID-19 all parts of the politico-healthcare establishment – NIH, CDC, WHO, FDA – have frequently been tone deaf in their pronouncements of “Trust the science” and “I represent science (with the corollary that those advocating unconventional approaches are quacks and cranks by my definition).”  Aside from immediate identification of SARS-CoV-2 as the agent of the COVID-19, very little true science informed the early responses.  to the pandemic.  This includes pronouncements from Dr. McCullough and his like-minded contrarians that “children are not at risk” and that “those who are infected with SARS-CoV-2 have permanent immunity.”  The first statement is outrageous.  The second statement was known to be untrue soon after the identification of respiratory disease caused by coronaviruses in the 1950s.

The response to COVID-19 by Dr. McCullough and his coworkers is not unlike that of the American Institute for Economic Research (AEIR) in their wonderfully named Great Barrington Declaration.  How could something with that name not be true?  Given that the AIER is a libertarian thinktank dating to Great Depression the text of the GBD could probably have been written by ChatGPT with the proper prompts.  But it is not true. [4]  But neither were the primary strategies promoted by the politico-healthcare establishment, which accelerated straight to mRNA vaccines and not much further.  And they had no excuse in not knowing that lasting immunity to coronaviruses is a chimera.  Thus, these vaccines prevent neither infection nor transmission.  But widespread and sustained use of effective masks would have worked to contain the pandemic, along with improvements in ventilation and air filtration.   During this time antivirals could have been developed, such as those that made AIDS a manageable disease for the vast majority of those who can afford the drugs.  That everyone who needs those drugs cannot get them is another matter altogether.

So, where do we go from here?  I honestly do not know, so let the discussion begin.  But while reading Everything is Predictable: How Bayesian Statistics Explain Our World by Tom Chivers, for a better understanding of the Reverend Bayes in my work, I found the kernel of an answer but not the solution in this quote:

We appear to be Bayesian machines (italics in original).  That’s true at a fairly high level: humans are rubbish at working out Bayes’ theorem formally, but the decisions we make in everyday life are pretty comparable to those that an ideal Bayesian reasoner would make.  Which, unfortunately doesn’t mean we all end up agreeing – if my prior beliefs are very different from yours, then the same evidence can lead us to entirely different conclusions, but sincere disagreements on apparently well-evidenced questions about the climate, or vaccines, or any number of other questions.

And we’re Bayesian at a deeper level, too.  Our brains, our perception, seem to work by predicting the world – prior probabilities – and updating those predictions with information from our senses: new data.  Our conscious experience of the world can be best described as our priors.  I predict, therefore I am.

Yes, we are Bayesian in our approach to the world around us.  Our problems lie in “sincere” disagreements that are not and “well-evidenced” questions for which the evidence is not disinterested.  Does the notional libertarian have a sincere disagreement with mask mandates on scientific and epidemiological grounds, or is he just irritated that his negative liberty is infringed upon when masks, which do work, are recommended?  Does the expressive leftist really believe the solution to urban blight in its many forms is to defund the police or is this idea simply performative nonsense?  In each of these examples the causes are much deeper than the reactions.

Regarding “well-evidenced” questions, the scientific support for the efficacy of COVID-19 mRNA vaccines was very thin from the beginning.  Whether they have been effective is a matter of definition, but they have not stopped transmission or infection, which is the general expectation of a vaccine.  The data showing that the current H5N1 virus is the result of laboratory “experiments” is likewise thin and the paper under discussion is tendentious in my view.  But what we can and must do is keep our wits about us and remain sincere and persuadable only by good, disinterested evidence, which is the only kind of scientific evidence.  SARS-CoV-2 was an unknown of sorts in November 2019, although it should not have been after the previous SARS and MERS outbreaks.  That we dropped them as a pressing scientific problem is unfortunate, not to mention ridiculous.  Influenza viruses are a known quantity.  What we must not do is forget that which we already know about them.  And remember what Maimonides said a thousand years ago, “You must accept the truth from whatever source it comes,” so long as the source does not have an axe to grind.

Even though Proximal Origin of Epidemic Highly Pathogenic Avian Influenza H5N1 Clade and Spread by Migratory Waterfowl clearly has an agenda, it is not impossible that serial passage of H5N1 has selected for mutations that make this variant more virulent.  A mechanism for how this variant could have origins on two continents requires “imagination,” but evolution can converge.  Still, preprints are just that, preliminary, in addition to being provisional, as are all scientific hypotheses.  One other tip when reading this literature: If the primary author is described as a Fellow (scroll down to lower left of page) of the organization taking responsibility for the paper but the sole contact information is his gmail address, that can be taken as a waving red flag.


Thanks to LS for suggesting this topic.  I take full responsibility for any and all deficiencies in the argument.



[1] I view Biomedicine as the often-corrupt offshoot of Biomedical Science, largely purveyed by Big Pharma, Big Ag, and Big Science, plus their marketing divisions, along with their antagonists on the Right and the Left.

[2] We have covered problems with peer review here, but the goal is to improve peer review, not discard it.  Sisyphus comes to mind.

[3] Some of whose work I have read.  Scruton was a conservative, apparently with a lowercase “c”, which means he was a Liberal with an uppercase “L”, who wrote in an interesting voice about modern philosophy.  His book on caring for the environment locally identifies where the individual should begin, even if it offers no general solution to the coming inconvenient apocalypse.  Sir Roger was also on the take for Big Tobacco.

[4] Note to the indispensable Matt Taibbi from a current subscriber and a reader since before the takedown of the Vampire Squid: One of the three primary authors of the Great Barrington Declaration, Jay Bhattacharya, MD-PhD, is not a scientist and he was never a physician.  His MD from Stanford means little other than adornment of his CV, because he never qualified to be a physician through residency and board certification.  Instead, he went from medical school directly to his PhD in the Stanford Department of Economics.  As I understand it, if he identifies himself as a physician or implies he is that kind of “doctor,” he is breaking the law in many states.

Print Friendly, PDF & Email


  1. hemeantwell

    Nice work, KLG, this introduction to libertarian science is very helpful. Re Bayesian stats, for a primer would you recommend the Chivers book you cited or something else?

    1. KLG

      I tend to think that Lord Rutherford was correct when he probably did not tell Max Planck (paraphrase), “If you have to use statistics, you should have designed a better experiment.” In my current day job I have to argue against the facile use of statistics as a predictive tool, so I am late to the game. Bayesian arguments are useful in molecular phylogenetics, which is where I got started later in the non-administrative part of my job. I am NOT a statistician, though!

      I will finish Chivers later this week. It is definitely a popular and valuable exposition. Probably a good place to start.

      On the history, use, and abuse of statistics:
      Bernoulli’s Fallacy: Statistical Illogic and the Crisis of Modern Science, Aubrey Clayton. This book was revelatory, to me.

      Bayesian statistics:
      Bayesian Statistics for Beginners: A Step-by-Step Approach, Therese M. Donovan and Ruth M. Mickey. Textbook but accessible.

      Statistics for People Who (Think They) Hate Statistics, Neil J. Salkind and Bruce B. Frey
      Includes SPSS, which I do not really need in my work.

      1. hemeantwell

        Thanks for that. I did some stat course work in graduate school, but that was in another far away galaxy and if Bayesian stats came up I probably confused it with aspirin.

  2. James E Keenan

    Does the expressive leftist really believe the solution to urban blight in its many forms is to defund the police or is this idea simply performative nonsense?

    Hmm, there’s a whole lot of interesting discussion to be opened up in that question.

  3. Samuel Conner

    I’m curious whether there are discernible differences between nations in the extent of these problems. China has become a research “powerhouse.” Might we hope that un- or less-biased research may be done there?

  4. Ignacio

    IMO, the paper titled “Proximal Origin of Epidemic Highly Pathogenic Avian Influenza H5N1 Clade and Spread by Migratory Waterfowl” doesn’t qualify as a scientific paper. It is an opinion thing and very speculative. It mixes data in the wildest way. It is, for instance, telling that the only reference I have checked with regards to genomic reassortments in virus populations in the wild, which are quite common and totally natural specially when a virus has spread as much as H5N1, are regarded in this “article” as possibly related with that stupidity called “gain of function research” that gained too much track with the Covid conspirational theorists. The paper has no experimental work done, no materials and methods, no nothing. At most it might be considered something like a “letter to the editor”. It shows nothing but speculations. The title starts with “proximal origin” and they show nothing that supports any proximal origin of any virus strain/variant. Not even phylogenetic analysis which would be the bare minimum.

  5. DavidZ

    The science and the lies purveyed by interested parties are a matter that I am not going to discuss.

    What I do think is the government should help those who want to protect themselves from airborne diseases. The best way would be to provide N95 respirators for free to everyone who wants one and their families. The government has mailed out free Covid tests to everyone who asked, they should do the same with N95 masks. A 3M N95 mask lasts around 1 week. so 52 + 10 extra per person. At retail they cost $3 each, with govt bulk buying – $1 each.

    This has some advantages:
    – Everyone who wants to protect themselves – elderly, immuno-compromised, the wise, customer interfacng workers – have no reason not to (aka cost).
    – More people wearing N95 masks, increases the amount of total protection available to the whole population.
    – Fewer Long Covid / healthcare problems for the whole population + lower healthcare costs + lower chances of new variants + higher economic activity.
    – Cost of N95 respirators is not a limiting factor (this is especially true for low wage customer interfacing workers making minimum wage, taxi drivers etc).
    – The more people wearing masks, the more it will become normalised and then it will become a habit during flu season etc.

    The second thing to do is fix the air-conditioning + filtering systems in Schools, hospitals, multi-family apartments and Public places.

    the first one is low hanging fruit.
    second one is more of a lift and can be done.

  6. tiebie66

    It is unfortunate that this piece, for me, came across as both pre-propaganda and a hit piece against McCullough-linked groups. Quoting Maimonides on the one hand “You must accept the truth from whatever source it comes,” while on the other cautioning against accepting the truth coming from certain directions seems rather contradictory. Should we reject the truth when it comes from those having an axe to grind? And is the implication here that McCullough-linked groups have axes to grind? So this whole piece says to be careful about believing material from the McCullough-linked groups but nary a sentence actually deals with the contents of the paper other than its objectives! No analysis of the contents of the paper nor an explanation of why it is tendentious. Just a “say so” statement to be accepted at face value: ”The data showing that the current H5N1 virus is the result of laboratory “experiments” is likewise thin and the paper under discussion is tendentious in my view.“ The lack of focus on the contents and the extensive focus on the context made the piece come across as an attack, not a critique.

    What I also found bothersome was the apparent arrogance implicit in “The second statement was known to be untrue soon after the identification of respiratory disease caused by coronaviruses in the 1950s.” In “SARS Cov-2, The World’s Gentlest Bioweapon” Leonardi points out that “sars cov 2 is enough of an anomaly and is sufficiently harmful physiologically to warrant avoidance and to be considered man-made” to presumably make a reliance on knowledge from the 1950s invalid due to its anomalous and man-made nature. The same arrogance of “knowing” was responsible for the fact that “Patients have been integral to recognizing long COVID as a legitimate condition” and not the medical establishment. (Fom the link to “The impacts of long COVID across OECD countries”).

    So, we have now been conditioned to approach the medical literature, not with an open mind and regardless of where it might come from, but with a pre-instilled bias against certain quarters. For me this piece is indeed very unfortunate as it has backfired on the author with the result that when reading future articles, I will start with doubts. At the same time I am aware that time and effort go into them and I fully acknowledge and respect that.

    1. Yves Smith Post author

      First, you seem unwilling to consider that some parties who opine about Covid are cranks or worse, operating in bad faith. McCullough is one. He has promoted the affirmatively dangerous and destructive notion of natural immunity, which is bogus with Covid, since any immunity is short short (on the order of 6-8 months) as to not be protective.

      Second, your other claims are false. From GM, who is a biomedical scientist and has been watching Covid evolution very closely (he was very early to call the outbreak of Omicron, when it was at 4 cases, and predict base on its number and placements of mutations that it would be highly immune/vaccine evasive and thus primed to spread very quickly). GM said via e-mail:

      That sort of naive extrapolation applies in the absence of selection.

      But viruses are under huge selective pressure all the time

      Third, as for your Leonardi quote, it does not prove what you said it proves. SARS-2 is still a coronoavirus, FFS, and even if it was lab-created, the process is effectively an acceleration of the same sort of selections you see in breeding, such as breeding chickens to have enormous breasts. A chicken is still a chicken despite breeding and SARS-2 is still a coronavirus.

    2. KLG

      Jeebus. What Georgi says.

      This has been a big part of my (slowly) ongoing research. Molecular clocks can work but they are NOT linear, in the sense that just any mutation in a coding sequence is the same as any other when it comes to random mutation. This has been known since the original clocks were constructed using globin sequences (hemoglobin subunits and the monomeric myoglobin) IIRC. At the protein level the first thing that jumped out in these comparisons is that some amino acids in a protein chain basically “never” change because they are essential for function. At the DNA level, because the genetic code* is degenerate (the second and third nucleotides in a triplet codon can be variable, this is somewhat hidden. But the ratio of silent and non-silent substitutions can be vary useful in identification of parts of a coding sequence that are under selective pressure. This was one of the signal discoveries of molecular evolution.

      Yes, TMI, but this is perhaps the too understated message of my post today: It does no good to tell these people out loud they are wrong (a scatalogical expression I would like to use has been suppressed). That just bounces off and gives these libertarians and other dumbasses more “immunity” to all criticism among their followers.

      Molecular evolution is complicated. Darn! Who would have thought that?

      *Genetic code is the nucleic acid (DNA/RNA) codon-amino acid (protein) key, not the complete sequence of the genome of any given virus or living organism (which is really only a lexicon and not a code for anything in the sense that lines of code in a program direct a computer to perform a task). I lost that battle long ago but not without a fight that still flares up from time to time.

      1. tiebie66

        Though I disagree with you both because I think that it is still too early to draw strong conclusions about sars cov 2 and associated countermeasures except masking, I do appreacite your responses.

  7. nyleta

    I must admit that the slow build up of the bird flu spread has me re-thinking my opinions about the lab origins of Covid. Perhaps it is just that one is mainly through domesticated animals and the other mainly through wild animals, such that the reporting is on a different timeline.

  8. CA

    KLG is terrific to read, and I always do so twice to be sure of my understanding. Thank you so much.

Comments are closed.