Gain-of-Function: Because Nature Was Taking Too Long
This sentence shouldn't sound normal:
“We made the pathogen more dangerous so we could learn how to stop it.”
Yet, that is the basic logic behind the most controversial edge of gain-of-function research.
In its cleanest, most defensible form, it is preparedness.
Scientists alter or adapt viruses and other pathogens to understand what mutations might make them spread better, infect new hosts, evade immunity, or become more severe. Then they use that knowledge to design vaccines, antivirals, surveillance systems, and diagnostics.
In its less charming form, it is humanity looking at evolution and saying, “Let us speed-run the apocalypse in a federally funded facility.”
The official language is all about “Enhanced potential pandemic pathogens.” “Dual-use research.” “Countermeasure development.” “Risk assessment.” “Threat reduction.” “Biosafety protocol.”
The facts are not that soothing.
Gain-of-function, or GOF, is not a single cartoon-villain experiment.
It's a category of work in which a pathogen gains a new or enhanced trait.
That trait can be higher transmissibility, increased virulence, altered host range, immune escape, better replication, or greater environmental stability.
Sometimes this is done by genetic engineering.
Sometimes by reassortment or recombination.
Sometimes it is done by serial passage, which is the wonderfully prehistoric method of repeatedly infecting cells or animals, harvesting what grows best, and letting laboratory evolution do the rest. Nature takes centuries. The lab has incubators and impatient grant cycles.
Serial passage is also part of legitimate vaccine history. Live attenuated vaccines have often been developed by repeatedly passing a virus through non-human cells, eggs, or animals until it becomes weaker in humans while still provoking immunity.
Vaccine production can also involve adapting viruses to grow efficiently in eggs or cell cultures.
The same general toolkit can therefore produce a weakened vaccine strain, a high-yield vaccine seed, a challenge strain for testing countermeasures, or a pathogen with more worrying properties.
The border between “vaccine development,” “risk assessment,” and “this seems like a bad idea” is not always obvious to the public, and sometimes not even cleanly agreed upon by the institutions funding the work.
The United States officially recognized this ambiguity.
In 2014, the U.S. government paused new federal funding for certain gain-of-function research that could be reasonably anticipated to increase pathogenicity or respiratory transmissibility in influenza, SARS, or MERS viruses. NIH lifted that pause in 2017 after the HHS P3CO framework was introduced, and in 2024 the U.S. issued a broader policy for oversight of dual-use research and pathogens with enhanced pandemic potential.
In 2025, the White House went further, ordering restrictions on federal funding for dangerous gain-of-function research while new standards were developed.
The US policy pendulum is funny: the Obama administration paused the dragon in 2014, the first Trump administration lifted the pause and added the 2017 P3CO leash, the Biden administration broadened oversight with its 2024 dual-use policy, and the second Trump administration swung hard the other way in 2025 with an executive order restricting dangerous gain-of-function work while ordering revisions to the previous framework.
The public is then told to trust the experts.
This would be easier if the experts, as a global class, had not spent the past few decades proving that even high-status laboratories can misplace pathogens, fail to inactivate samples, infect workers, contaminate shipments, leak virus through plumbing, and discover smallpox in a forgotten freezer like an old lasagna.
Documented incidents
The documented record is not complete, because reporting is inconsistent across countries, institutions, military programs, private contractors, public-health labs, and university labs.
But still, interesting enough.
A 2024 Lancet Microbe scoping review found 309 laboratory-acquired infections in 94 reports involving 51 pathogens between 2000 and 2021.
Eight of those infections were fatal.
The same review identified 16 accidental pathogen escapes from laboratory settings during that period.
That does not mean 16 pandemics. It means 16 documented incidents where pathogens got beyond where they were supposed to be, plus hundreds of infections acquired by people working with or around them.
In 2003 and 2004, SARS escaped laboratories more than once after the original SARS outbreak had been contained. Singapore had a laboratory-acquired SARS infection in 2003. Taiwan had a SARS lab infection in December 2003. Beijing had a 2004 outbreak linked to the National Institute of Virology, where experiments with live and inactivated SARS coronavirus had been carried out.
WHO reported that two researchers developed SARS in late March and mid-April 2004, the institute was closed, and the outbreak infected nine people and killed one.
In 2005, live H2N2 influenza - the 1957 pandemic strain - was accidentally sent as part of proficiency testing to 3,747 laboratories in 18 countries. Most recipients destroyed the samples once the problem was recognized. No outbreak followed, but “we mailed a pandemic flu strain around the world by mistake” is not a sentence any civilization should have to file under “administrative incident.”
In 2007, foot-and-mouth disease virus leaked from the Pirbright site in the United Kingdom.
Official investigations found the likely route involved faulty drainage pipework, contaminated soil, and movement from the site to nearby farms. Livestock were infected and animals were culled.
Not a human pandemic, but it was a real-world release from a high-containment animal-disease research and vaccine site.
In 2014, the U.S. Centers for Disease Control and Prevention had a cluster of alarming lab safety failures.
CDC described potential anthrax exposure after improper inactivation procedures, an incident involving a dangerous flu virus, and the discovery of vials labeled smallpox in a storage room on the NIH campus.
The smallpox discovery later involved six variola vials found among hundreds of unclaimed vials in an FDA-occupied building.
In 2015, the U.S. Army’s Dugway Proving Ground was found to have mistakenly shipped live anthrax samples that were supposed to be inactivated. Reports later described live anthrax being sent to more than 80 laboratories in seven countries, and a Department of Defense review called it a systemic failure.
There are older cases too.
Laboratory-acquired infections with dangerous viruses such as Marburg and Sabia have occurred.
Sabia virus infected a researcher at Yale in 1994 after a centrifuge accident.
Marburg infections occurred in Soviet/Russian high-containment research contexts.
These events did not become global outbreaks, but they prove the principle.
Newer tallies since the Lancet review push the total documented lab-acquired infections closer to 435 when you include data stretching back to the mid-1970s - mostly bacteria like Brucella in China, but viruses keep showing up in the mix.
Then there are the influenza experiments that made the whole modern debate much louder.
In 2012, Ron Fouchier’s group reported that genetically modified H5N1 avian influenza acquired mutations during passage in ferrets and became airborne transmissible between ferrets.
Yoshihiro Kawaoka’s group produced a reassortant H5 virus with mutations in hemagglutinin and the remaining genes from 2009 H1N1 that could spread by droplets in ferrets.
The stated rationale was pandemic preparedness: learn what mutations could make H5 viruses more transmissible in mammals, then improve surveillance and countermeasures.
The obvious objection was also pandemic preparedness: perhaps do not create airborne-transmissible mammalian H5 viruses.
Coronaviruses are only one part of this story, but they are the part that made the public notice.
No name looms larger here than Ralph Baric at the University of North Carolina.
Baric is the reverse-genetics wizard who shines in coronavirus research.
His lab spent years building infectious clones - basically, the full genetic blueprint of a coronavirus copied as DNA fragments that could be cut, pasted, and “rescued” back into live virus in a dish.
Think Lego for viruses: swap any piece and see what happens.
The headline act came in his 2015 Nature Medicine paper.
Baric’s team (led by Vineet Menachery, with Shi Zhengli - the bat woman - from Wuhan as co-author) took the spike gene from a bat coronavirus called SHC014 - sequenced from Chinese horseshoe bats but never grown in the lab - and stitched it onto the backbone of a mouse-adapted SARS-CoV strain (MA15) that Baric had already engineered.
They also rebuilt the full-length SHC014 virus itself.
This was classic gain-of-function work: a chimeric virus that had never existed in nature.
What did it do?
In primary human airway epithelial cells (the real stuff lining your lungs), it replicated like a champ - titers matching the 2003 SARS epidemic strain. It used the human ACE2 receptor efficiently.
In mice, it infected the lungs and caused weight loss and inflammation, but it was actually less lethal than the pure mouse-adapted SARS parent.
Baric has always stressed this point: in the mouse model, it was more “loss of function” than gain.
Still, the paper showed that some bat viruses sitting in caves were already one spike swap away from efficient human-cell infection.
They tested antibodies and antivirals against it - classic risk-assessment framing.
The work was reviewed and approved under the rules at the time; it mostly happened at UNC’s BSL-3 lab, not Wuhan (though Shi provided the bat sequence).
Baric deliberately left the exact chimera sequence out of the published paper to make replication harder without permission. He followed up with similar chimeras in later papers, including a 2016 PNAS study on WIV1-CoV.
All of it fed straight into the mRNA vaccine platforms and monoclonal antibodies.
Critics saw the dual-use problem in neon lights: you just published the roadmap for building a human-infecting coronavirus in a dish.
Peter Daszak and EcoHealth Alliance were the matchmakers who made much of this possible.
EcoHealth, Daszak’s nonprofit, received roughly $3.5 million from NIH/NIAID for bat coronavirus work.
About $600,000 of that flowed as a sub-award to the Wuhan Institute of Virology for field sampling, sequencing, and some of the lab experiments.
https://media.nature.com/lw767/magazine-assets/d41586-020-02473-4/d41586-020-02473-4_18305404.jpg
Daszak’s group collected the viruses in the field, Baric’s lab supplied the engineering know-how, and Shi’s team ran the tests.
Later investigations found EcoHealth was late on reports, failed to promptly flag a chimeric virus that grew better than expected in mice, and the whole arrangement became a fog of grant language and finger-pointing.
By early 2025, HHS had debarred both EcoHealth and Daszak from federal funding for five years over those oversight failures.
Daszak still calls it essential risk assessment; Baric has pushed for tighter global biosafety standards while defending the science as necessary.
Anthony Fauci enters the story not as a man in a cape stirring a cauldron, but as the long-serving head of NIAID - the funding. NIAID money supported the broader bat coronavirus research ecosystem.
Fauci testified that NIH did not fund gain-of-function at Wuhan, but the details of what counted under shifting definitions became a political and scientific firestorm.
Ukraine
Ukraine belongs here, but not as a lazy copy-paste from coronavirus arguments.
The documented U.S.-Ukraine biological programs were Defense Threat Reduction Agency and Cooperative Threat Reduction activities, beginning after the Soviet collapse and formalized in Ukraine in the 2000s.
The stated purpose was to secure dangerous pathogens, improve biosafety and biosecurity, support disease surveillance, and convert former Soviet biological infrastructure toward public-health use.
DTRA’s own fact sheets say the program worked with Ukraine to reduce the chance that pathogens could be misused, stolen, or accidentally released - and similar work had even occurred with Russia until 2014.
These labs handled endemic stuff like anthrax, plague, tularemia, and avian flu surveillance.
Defensive public-health infrastructure in a war zone carries real risks, and “trust us, the defense agency is helping with pathogen surveillance” sounds different to different ears.
~~~
China had the reservoir access.
Southern China and nearby regions contain horseshoe bats carrying SARS-like coronaviruses, and Wuhan had a major coronavirus research center with Shi Zhengli’s group and large bat-virus collections.
The United States had money, methods, and institutional interest.
Europe had its own high-risk influenza debates.
Russia had the Soviet legacy and a habit of accusing others of the sort of biological darkness its own predecessor state knew intimately.
Australia had a modern public-health lab that still managed to lose track of dangerous virus samples.
This is the part where the article stops being a national scandal and becomes a species problem.
The Queensland incident:
In 2021, during a freezer-transfer mess that was later uncovered, hundreds of vials containing Hendra virus (57% fatality in humans), lyssavirus (basically rabies), and hantavirus samples became unaccounted for at Queensland’s Public Health Virology Laboratory.
Queensland Health later reported that the samples were unlikely to have been lost or stolen, were unaccounted for because of poor record-keeping, and caused no risk or harm to staff or the broader community.
The mice
The mouse is a brave civil servant.
It expresses the receptor we ask it to express.
It dies in measurable ways.
It generates graphs.
It has never once objected to being used as a proxy for all humanity.
But mice are not humans.
Even “humanized” hACE2 mice - engineered to carry the human receptor for SARS-like viruses - are a clever hack, not tiny people with tails. Wild-type mice don’t support efficient infection at all.
The transgenic ones often over-express the receptor in weird places (hello, brain encephalitis that doesn’t mirror typical human disease).
Their immune systems, lungs, metabolism, and lifespans are nothing like ours.
They miss the messy human reality of obesity, diabetes, age, co-infections, pollution, or bad ventilation.
Ferrets are better for some respiratory transmission questions, hamsters for certain coronavirus models, non-human primates closer still - and none are us.
A virus that behaves one way in a mouse cage may behave differently in a subway, a school, a hospital, a prison, a factory, or a care home.
The vaccine-development defense is also real and incomplete.
Yes, studying dangerous pathogens can help identify risk markers, improve diagnostics, build vaccine platforms, test antivirals, and prepare for natural spillovers.
Yes, some serial passage and viral engineering techniques are part of legitimate vaccine science.
Yes, surveillance of zoonotic viruses can matter.
Yes, not every frightening experiment is irresponsible.
The problem is that “can help” is not the same as “is worth doing,” and “done for preparedness” is not the same as “safe enough for civilization.”
The pandemic preparedness industry often speaks as if the benefit is concrete and the risk is theoretical.
The documented incident record says otherwise.
The benefit is often probabilistic: maybe this mutation map helps surveillance; maybe this animal model improves a vaccine; maybe this challenge strain helps screen an antiviral; maybe this dataset predicts the next spillover.
The risk is also probabilistic, but sometimes it comes with dates, facilities, infected workers, dead animals, destroyed shipments, locked-down contacts, and reports that begin with “an investigation has been launched.”
The financing structure? Dark af.
The work is funded through national health agencies, defense agencies, universities, NGOs, private contractors, international collaborations, and subawards. NIH/NIAID money supported U.S. academic and NGO-led pathogen research.
DTRA money supported biological threat reduction abroad.
European and Asian governments funded their own high-containment and influenza research ecosystems. Vaccine manufacturers and public-health labs handled production, diagnostics, and surveillance. Everyone can point to a different line in the budget and say, “That part was not us.”
Fauci, Baric, Daszak, Shi, Fouchier, Kawaoka, DTRA, NIH, CDC, WIV, Pirbright, Dugway, Queensland Health are not interchangeable villains. They are playing with potential pandemics.
“Interesting” is a cursed word in virology.
The nonpolitical conclusion is not “America bad,” “China bad,” “Ukraine bad,” “Russia right,” or “scientists evil.”
The conclusion is that every major power wants the benefits of high-risk biology while minimizing the visible liability.
The United States hides behind definitions and funding layers.
China hides behind opacity and state control.
Russia hides behind propaganda while carrying the historical baggage of Soviet biological weapons.
Europe hides behind committees and acronyms.
Ukraine shows how even defensive public-health labs become terrifying in a war zone.
Australia shows that wealthy, modern systems can still lose track of samples.
The private and academic sectors hide behind publication prestige, intellectual property, and the comforting phrase “standard practice.”
The public should not have to choose between anti-science paranoia and institutional obedience.
The sane position is harsher and more boring: publish the funding trails, disclose the pathogen inventories, require independent adversarial review, make lab incidents mandatory and public, separate vaccine production from avoidable enhancement work wherever possible, stop outsourcing accountability through subawards, and treat “we followed protocol” as the beginning of the conversation.
The next lab incident will not care which flag was outside the building, or whether the work was called vaccine development, threat reduction, biodefense, surveillance, reverse genetics, dual-use research, or preparedness.
It will only care whether the door, vial, drain, freezer, courier, centrifuge, waste system, inactivation protocol, animal handler, inventory spreadsheet, and chain of command all worked perfectly on the same day.
And if the entire safety case depends on that many humans not being human, then perhaps accountability was taking too long.
mRNA "vaccines"
The vaccine-development defense is murky.
The crown jewel of recent countermeasure technology - mRNA platforms - carries its own uncomfortable dangers.
The lipid nanoparticle (LNP) delivery system, essential for getting mRNA inside human cells, often includes components like polyethylene glycol (PEG) that are known allergens capable of triggering anaphylaxis in a subset of people.
Manufacturing “Process 2” - the scaled-up version used for mass deployment - differed significantly from the cleaner “Process 1” used in clinical trials, with reports of higher levels of residual DNA fragments, SV40 promoter elements, and endotoxin contamination.
What began as an elegant but questionable idea (teach the body to make its own spike protein) became, in practice for many, a rushed “poojab” rollout with unprecedented scale, novel mechanisms, and safety signals that regulators and manufacturers have been slow to address transparently.
The same toolkit that flowed from GOF-style research into rapid-response vaccines also accelerated questions about long-term effects, immune imprinting, and whether we traded one set of risks for another.
~~
The media circus only makes the fog thicker.
Even Serious outlets often swing between breathless “lab-leak bombshell” headlines one week and dismissive “debunked conspiracy” pieces the next, chasing traffic, ad revenue, and the right expert quote that fits the prevailing narrative.
The less serious players crank the volume even higher: dramatic music, red arrows, and claims that every lab incident is proof of an active bioweapon program.
The independent voices and conspiracy corners - figures like Sasha Latypova with her detailed but often overreaching documents on DOD contracts and pharma manufacturing, or Jane Ruby hammering “poison shot” narratives on podcasts and Substack.
Some of their points land because the official story has holes; others spin into full fanfiction where every bat virus, every sub-award, and every mouse model becomes part of a grand depopulation plot with secret cabals and alien tech.
The Covid lesson
Opaque buying, redacted contracts, liability shields, and state pressure for forced jabs were never going to build trust.
The von der Leyen/Pfizer SMS scandal didn't help either - when billions in public money are involved, “the messages are unavailable” is not a transparency policy.
Add freedom of choice and bodily sovereignty. People do not stop owning their bodies because a government says so.
Informed consent matters.
Society is still living with the fallout: not only from the vaccines themselves, but from the mandates, exclusions, censorship fights, broken trust, family splits, professional pressure, and unanswered questions around safety signals, procurement, liability, and decision-making.
It does prove that the people demanding public trust were often careless, arrogant, and far too comfortable hiding the receipts.
Those events are not fully cleared up.
They should not be memory-holed just because the news cycle moved on.
~~
For most people who don’t have time (or the training) to read 2015 Nature Medicine supplements, Baric’s reverse-genetics methods, or the latest Lancet scoping review on lab incidents, it’s exhausting.
The best lies are those with a grain of truth.
Real lab accidents have happened, real risks exist, and real shortcuts have been taken in the name of “preparedness.”
That's no excuse for hysteria, doomsday fanfiction, or spreading panic.
The sane response is - slow down, read the primary papers when you can, cross-check claims, and use tools like AI to translate the jargon.
Don’t outsource your thinking to any side.
You might also like
