What do pythons, house cats, and chicken eggs have in common? Apparently, they're all quietly upstaging the pharmaceutical industry. This week, we've got a snake metabolite coming for Ozempic's lunch (literally), your cat's tumor DNA looking suspiciously like yours, a study confirming what we all suspected about scientific humor, a student in a mesh suit generating more mosquito flight data than all of history combined, and a startup replacing billion-dollar bioreactors with poultry. Let's go.

RESEARCH

You know how, after Thanksgiving dinner, you just want to lie on the couch for three hours? Now imagine doing that for 18 months. Burmese pythons can swallow prey equal to their body weight and then go over a year without eating, all while keeping their muscles and heart in perfect shape. Researchers at CU Boulder, Stanford, and Baylor wanted to know how.

Turns out, the secret is in the blood. The team, led by Leslie Leinwand, a distinguished professor who's been studying pythons for two decades, measured blood from ball and Burmese pythons immediately after feeding. They found 208 metabolites that spiked significantly, but one molecule stole the show: para-tyramine-O-sulfate (pTOS), which soared over 1,000-fold. For context, in humans, pTOS rises a modest 2 to 5 times after a meal.

When the team gave pTOS to obese mice, the results were promising: the mice ate significantly less, losing 9% of body weight over 28 days compared to controls. The molecule works by activating neurons in the ventromedial hypothalamus, the brain's satiety center, and it's produced by gut bacteria from dietary tyrosine. Crucially, unlike GLP-1 drugs like Ozempic, pTOS didn't cause GI problems, muscle loss, or energy drops.

This is the second time reptiles have gifted us weight-loss leads (GLP-1 drugs were inspired by Gila monster venom). Leinwand and colleagues have already formed a company, Arkana Therapeutics, to develop the findings. Though to be fair, pTOS hasn't been tested in humans yet, and its appetite-suppressing effect may be reduced in people with prediabetes or type 2 diabetes. So maybe don't cancel your Ozempic prescription just yet.

RESEARCH

If you've ever suspected your cat was secretly plotting something, congratulations: it turns out they've been holding onto cancer genomics data this whole time. An international team has just produced the first large-scale genetic map of feline cancer, and the results are uncomfortably familiar.

Published in Science, the study sequenced tumors and healthy tissue from 493 pet cats across five countries, covering 13 different cancer types and screening roughly 1,000 human cancer gene orthologs. The team, led by the Wellcome Sanger Institute with collaborators at the University of Guelph, University of Bern, and Cornell, found that TP53 was the most frequently mutated gene at 33% of feline tumors (a rate that mirrors human cancers). They identified 31 driver genes, mutational signatures, and germline variants in total.

The standout finding? FBXW7 was altered in more than half of feline mammary tumors, a gene linked to worse prognosis in human breast cancer. When the team tested chemotherapy drugs on feline mammary carcinoma tissue in culture, certain treatments showed higher efficacy against FBXW7-mutant tumors. These are tissue culture results, not clinical trials, but it's proof of concept for cross-species therapeutic strategies.

Because cats share our living environments (and presumably our questionable lifestyle choices, at least partly), they're exposed to similar environmental cancer risks. "This confirms that the domestic cat is not just a beloved pet, but a vital partner in the fight against cancer," said co-author Latasha Ludwig of Cornell. The study's data is now an open resource for researchers. Precision feline oncology is apparently a real field now, and honestly, that might be the most on-brand thing cats have ever done.

RESEARCH

We've all sat through that conference talk where someone tries a joke and gets back the sound of 200 people politely refusing to make eye contact. Well, a team of researchers decided to actually quantify that misery, because of course they did.

Stefano Mammola, an ecologist at the Italian National Research Council, teamed up with Victoria Stout, an environmental scientist at CU Boulder who moonlights as a stand-up comedian. Together with six other co-authors, they attended 14 biology conferences and tracked humor across 531 talks between 2022 and 2024. The resulting paper, published in Proceedings of the Royal Society B, is titled "Statistically significant chuckles," which is already funnier than most of the jokes they recorded.

The findings: about 42% of speakers didn't attempt any humor at all. Of the 870 jokes that were attempted, 66% earned only polite chuckles, and just 9% got whole-room laughter. The most used? Technical snafus like slides malfunctioning. Likely unintentional, but we get it. (Nothing builds community like watching someone else's laptop betray them. Because we’ve all been there)

There's a less funny dimension too. Male speakers told about 0.35 more jokes per talk, and both male and native English speakers had a 10% higher probability of eliciting laughter. The authors argue that humor in academia remains "a privilege" shaped by social dynamics and risk tolerance. Until the field reckons with those biases, conference comedy will remain an uneven playing field.

RESEARCH

Somewhere in a Georgia Tech lab, an undergraduate named Christopher Zuo put on a mesh suit, walked into a room with 100 hungry female Aedes aegypti mosquitoes, and generated more mosquito flight data than had previously been measured in human history. His reward? Getting bitten through the mesh. Science is a calling.

The study, a collaboration between MIT and Georgia Tech, published in Science Advances, produced the first 3D model of mosquito flight behavior. Using a Photonic Sentry camera that tracks hundreds of insects at 100 frames per second at 5mm resolution, the team selected 20 million mosquito positions and speeds and used Bayesian inference to build a predictive model.

They identified three distinct flight patterns: "fly-bys" when mosquitoes can only see a target (quick dive in, quick exit), "double-takes" when they can only smell CO2 (slowing down, flitting back and forth), and "orbiting" when both cues are present (circling like a tiny, disease-carrying shark). Mosquitoes, it turns out, don't swarm because they follow each other. Each independently picks up on cues and ends up in the same place. "It's like a crowded bar," said Georgia Tech's David Hu.

For context, this matters because mosquitoes kill over 770,000 people a year. The model could help design traps with better multisensory lures. Also worth noting: before mesh suits, scientists in the 1980s apparently conducted "bite studies" by stripping to their underwear. Progress.

NEWS

The current gold standard for manufacturing biologic drugs involves Chinese hamster ovary (CHO) cells, a system that traces back to a mid-century laboratory accident (we are sensationalizing here) rather than any deliberate engineering. Neion Bio thinks that's absurd, and they may have a point.

The New York-based startup emerged from stealth on March 26 with $11 million in financing and a commercial deal already signed. Founded in 2024 by CEO Dimi Kellari and CTO Sam Levin, the company's Raptor™ platform works by engineering avian primordial germ cells to carry therapeutic genes, introducing them into developing embryos, and harvesting recombinant proteins from the resulting hens' egg whites. Their CSO, Sven Bocklandt, previously directed genetic engineering at Colossal Biosciences (the de-extinction people, because of course).

The pitch: eggs are self-contained, naturally sterile, and evolution has spent 200 million years optimizing them for protein production. Levin estimates that fewer than 4,000 hens could meet global demand for a blockbuster antibody like Humira at what the company claims is one-hundredth the operating cost of an equivalent CHO facility. Their first commercial deal is a co-development and supply agreement for up to three monoclonal antibodies with an unnamed major pharma company, including upfront payments, milestones, and profit sharing.

To be fair, these are company claims, and the pharma partner remains unnamed. But if the platform delivers, the days of CHO cell dominance might actually be numbered. Egg-based biomanufacturing: nature's original bioreactor, finally getting its due.

This week's stories have a theme, and it's this: nature has been solving our hardest problems for millions of years, and we keep being surprised about it. Pythons figured out appetite regulation. Cats have been carrying cancer genomic data on their chromosomes. Mosquitoes solved multisensory navigation. Chickens perfected protein manufacturing. And scientists? Scientists apparently still can't figure out how to land a joke at a conference.

If any of these stories made you do more than a polite chuckle (that puts us in the top 33%, apparently), forward this to someone who'd appreciate it. And if you've got thoughts on which animal will upstage pharma next, hit reply. Our money's on tardigrades.

Keep questioning everything (especially hamster-based manufacturing paradigms),

Prateek & Jere

P.S. Eight co-authors to prove scientists aren't funny. The field of humor research is apparently immune to irony.