Evidence over ideology. Explaining science in context, exposing harmful misinformation, sharing references—and posting reminders the world is beautiful.
Joined August 2011
- Tweets114K
- Following6.4K
- Followers12.1K
- Likes109K
Pinned Tweet
Most people know Dr. Peter Hotez from his television appearances during COVID, his defense of vaccination, and the relentless attacks directed at him on social media.
I know Peter as a friend. And I want to tell you about another part of his life that rarely enters those arguments.
For more than four decades, Peter has devoted much of his scientific career to diseases that afflict some of the poorest people on Earth. Diseases that cause enormous suffering but attract relatively little commercial investment because the people who need treatment often cannot afford to pay for it.
Hookworm. Schistosomiasis. Chagas disease. Leishmaniasis.
These are hardly household names. Their victims rarely make headlines, and the scientists working on them seldom become celebrities.
Peter chose this field long before COVID, long before the cable news appearances, and long before strangers began calling him a pharmaceutical industry shill.
His career deserves a closer look.
Because the work tells a story that is almost entirely absent from the accusations made against him.
1/ I’ve seen people on X claim that “mRNA has no place in the human body.”
That’s an extraordinary misunderstanding of one of the most fundamental processes in biology.
Your body produces messenger RNA (mRNA) continuously. It has been doing so since the earliest stages of your development.
DNA contains the genetic instructions for making proteins. mRNA carries copies of those instructions to ribosomes, the cellular machinery that assembles proteins.
Those proteins include enzymes that regulate metabolism, receptors that allow cells to communicate, structural components of muscle, and molecules essential to immune function.
Without mRNA, your cells couldn’t manufacture the proteins required to maintain themselves.
This isn’t some recent discovery associated with COVID vaccines. It’s foundational molecular biology, established decades before mRNA vaccines existed.
🤖 Made with AI
2/ Consider a thought experiment.
Imagine every molecule of mRNA suddenly disappeared from your body, and your cells could no longer replace it.
Protein synthesis would effectively cease. Ribosomes would remain, DNA would still contain its genetic instructions, but the messages connecting those instructions to protein production would be gone.
Importantly, your existing proteins wouldn’t instantly disappear. Different proteins survive for different lengths of time, from minutes to days or considerably longer.
But proteins are continually degraded, damaged, and replaced.
Without new protein production, essential enzymes and regulatory proteins would progressively disappear. Cellular repair would fail. Metabolism, electrical signaling, muscle contraction, and immune function would deteriorate.
Eventually, vital organs could no longer sustain life.
The precise timeline would vary across tissues. The outcome of permanently eliminating mRNA-dependent protein production would not.
No mRNA. No sustained protein synthesis. No continued human life.
3/ There’s another important distinction frequently lost in discussions of mRNA vaccines.
Your cells naturally manufacture mRNA, use it to produce proteins, and degrade it through normal cellular processes. This happens constantly throughout your life.
An mRNA vaccine uses that existing biological machinery by delivering a manufactured RNA message encoding a particular antigen. Some cells translate that message into protein, allowing the immune system to recognize the antigen and develop a response.
Naturally occurring mRNA and vaccine mRNA aren’t identical in their sequences, formulation, or biological context. Nor does the fact that mRNA is essential to life automatically establish that every mRNA-based medical product is safe.
Safety still requires evidence specific to the product, dose, delivery system, and clinical outcomes.
But claiming that mRNA itself has no place in the human body?
That’s not a legitimate scientific objection to vaccination. It’s a misunderstanding of how human cells function.
H5N1 has crossed from birds into dairy cattle, established sustained transmission, and begun adapting to a new mammalian host.
What can the virus’s genome tell us about how this happened—and what it might mean for human health?
Using Nextstrain’s genomic surveillance data, this short thread explores what we know, what remains uncertain, and why monitoring viral evolution matters.
🧵👇
2/3 — The evolutionary significance of H5N1 in cattle.
Perhaps the most consequential finding is that H5N1 has demonstrated an ability to establish transmission in a mammalian livestock population.
And it has crossed that species barrier more than once.
The original cattle outbreak involved genotype B3.13. In 2025, genomic surveillance identified genetically distinct D1.1 viruses in dairy cattle in Nevada and Arizona, representing additional, independent introductions from birds.
Researchers have also identified mutations associated with improved replication in mammalian cells.
One example is PB2 M631L, a polymerase mutation found in the cattle-associated B3.13 lineage. Experimental studies show that this mutation can improve viral replication in bovine cells and certain human airway cell models.
That matters biologically. But it requires careful interpretation.
A virus becoming better at replicating inside mammalian cells is not necessarily becoming better at transmitting between humans.
Successful human respiratory transmission involves additional biological requirements, including receptor interactions, infection dynamics, and the ability to spread efficiently between people.
Genomics can identify potentially important adaptations. Laboratory and epidemiological evidence must establish what those adaptations actually accomplish.
The distinction is fundamental to understanding pandemic risk.
3/3 — What genomic surveillance tells us about the risk to humans.
The continued evolution of H5N1 in cattle deserves serious scientific attention.
A virus circulating in a large mammalian population has repeated opportunities to acquire mutations, undergo selection, encounter other influenza viruses, and occasionally cross into humans.
Human infections associated with exposure to infected dairy cattle have already occurred.
Yet an essential distinction remains: we have evidence of zoonotic transmission, but no established evidence of sustained human-to-human transmission.
Nor can we look at a phylogenetic tree, identify a handful of mutations, and conclude that a pandemic is approaching.
Genomic surveillance has limitations. Sampling is uneven, many infections go unsequenced, and genetic similarity alone cannot establish exactly how transmission occurred. Mutations that appear important require functional investigation.
The strongest assessments combine genomic data with epidemiology, experimental virology, animal surveillance, and careful monitoring of human infections.
That is why continued surveillance matters.
We are watching a virus adapt to an ecological environment that was previously unusual for it. Some of those changes are scientifically significant. Their implications for human transmission remain uncertain.
Genomic surveillance is an early-warning system, not a crystal ball.
Its purpose is to help us recognize consequential evolutionary changes while there is still an opportunity to respond.
UPDATE: The controversial hepatitis B vaccine trial backed by RFK Jr. will NOT proceed.
Guinea-Bissau has formally rejected the proposed U.S.-funded study involving 14,000 newborns, following months of international criticism over its scientific justification and ethical implications.
The trial would have randomly assigned half the infants to receive the hepatitis B vaccine at birth, while the remaining half would receive it at six weeks, consistent with the country’s existing vaccination schedule.
The investigators wanted to examine possible “nonspecific effects” of vaccination, including health outcomes unrelated to hepatitis B infection.
But there was a fundamental ethical problem.
We already know that administering hepatitis B vaccine shortly after birth substantially reduces mother-to-child transmission. The WHO estimates that the birth dose prevents 70–95% of these infections. And approximately 90% of infants infected around birth develop chronic hepatitis B, placing them at lifelong risk of cirrhosis and liver cancer.
The researchers argued that no child would receive less than the existing national standard of care. That distinction matters. But the WHO countered that resource constraints do not justify withholding a proven preventive intervention in a research experiment, particularly without a compelling scientific rationale.
On October 16, Guinea-Bissau’s government formally rejected the study. Researchers subsequently confirmed that it would not proceed. No children had been enrolled.
This is about something larger than RFK Jr. or any particular vaccine.
Scientific curiosity is essential. Investigating unexpected effects of medical interventions is legitimate. But research involving human participants must satisfy ethical standards that extend beyond whether an interesting question remains unanswered.
The burden is on investigators to demonstrate that a study’s scientific value justifies its risks—not on vulnerable participants to bear those risks simply because researchers want answers.
That principle deserves defending, regardless of who funds the research.
A Controversial Hepatitis B Vaccine Study Championed by RFK Jr. Is Axed After Widespread Backlash rollingstone.com/politics/po… via @RollingStone @KatherineEban
IntegralAnswers retweeted
There is something really concerning (and disappointing) going on in the US right now.
A dangerous and preventable situation.
Let me explain.
nextstrain.org/avian-flu/h5n…
1/
IntegralAnswers retweeted
CDC Sent Just One Disease Detective to Pennsylvania to Investigate Their Measles Outbreak "This is unlike previous outbreaks, in which EIS typically sent at least two disease detectives AND a team of CDC workers to help them." vaxopedia.org/2026/10/22/cdc… via @AboutPediatrics
The biggest lie about information | Yuval Noah Harari youtube.com/shorts/4A9YvaJbx… via @YouTube
IntegralAnswers retweeted
US measles cases hit 4,588, more than twice last year’s total cidrap.umn.edu/measles/us-me… via @cvdall @LizSzabo @CIDRAP
A recent interview of Bill Gates by Ezra Klein inspired me to take a closer look at some of the extraordinary warnings Gates is making about artificial intelligence.
At one point, Gates declared:
“This makes nuclear weapons look like nothing.”
He argues that AI has already crossed dangerous thresholds in cybersecurity and biological research, while governments and society remain inadequately prepared for the consequences.
These are serious claims, coming from someone with decades of experience in technology and global health.
But what has AI actually demonstrated? Which threats are credible? And where does the evidence end and speculation begin?
I examined Gates’s arguments alongside the available research, including evidence that supports—and challenges—his conclusions.
A thread. 🧵
🤖 Made with AI
11/ Gates’s warning deserves serious attention. But taking a warning seriously does not require accepting every prediction attached to it.
The evidence already establishes that advanced AI systems possess capabilities with important implications for cybersecurity, scientific research, and economic activity.
Some dangerous applications are credible enough to justify precautionary measures now.
Other consequences remain uncertain, particularly predictions involving catastrophic biological attacks, permanent mass unemployment, or autonomous systems escaping human control.
The appropriate response is neither technological fatalism nor unquestioning optimism.
It is independent evaluation, transparent evidence, proportionate safeguards, and institutions capable of adapting as the technology changes.
Perhaps the most important question isn’t whether Gates is right about every danger he foresees.
It’s whether we can build a system of governance capable of recognizing which dangers are real before we discover the answer through catastrophe.
That is a question worth confronting now.
Selected references and further reading:
• International AI Safety Report (2026) — frontier capabilities, misuse risks, uncertainty, and safeguards.
• Anthropic, Project Glasswing / Claude Mythos Preview (2026) — AI-assisted cybersecurity vulnerability discovery.
• UK AI Security Institute, Frontier AI Trends — empirical evaluations of advanced AI capabilities.
• ILO–NASK, Generative AI and Jobs: A Refined Global Index of Occupational Exposure (2025).
• European Investment Bank, AI Adoption, Productivity and Employment (2026).
These sources help distinguish demonstrated capabilities from plausible threats and uncertain predictions. Gates’s interview is the primary source for the arguments examined in this thread.