Welcome back hot friends.

For most of its history, vaccine development has been a reactive business: a virus appears, scientists scramble, and an injection arrives just in time for the variant it targetted to have already moved on. DIOSynVax decided to skip the chase entirely. They fed an AI every Sarbeco coronavirus genome they could find, including the bat viruses still minding their own business in caves, and asked it to design a single "super-antigen" that captures what the whole family has in common rather than any one member. 

The result, their vaccine has now passed its first human trial in 39 volunteers: safe, no significant side effects and immune responses against COVID, plus some bat viruses that haven't even bothered us yet. It is, the team notes, the first vaccine whose active ingredient was dreamed up entirely in a computer to reach human testing.

🧐 What's in it for me? For all the talk of AI doing admin, here it has achieved something genuinely consequential: designed a vaccine that works in people. The promise is a vaccine you don't renew every autumn like a tax disc, one built to cover tomorrow's variants and possibly an entire viral family at once. It's early (Phase 2 is next), but the same trick is already being pointed at flu and Ebola.

💵 Out of the Lab: If antigens can be designed by software rather than discovered by trial and error, the bottleneck in vaccine-making shifts from the lab bench to the GPU cluster, which is a very different competitive landscape.

Salamanders can regrow a lost limb. Humans grow a scar and call it a day. The difference, biologists have assumed since roughly Aristotle, is that we simply lack the machinery. However, some researchers suspected that the machinery is there, just switched off, and they appear to have found the switch. 

After amputating mouse digits and letting them heal (mean, we know), they hit the wound with a chemical signal, then a second a few days later. The first talked the scar-making cells out of their usual job and into building a blastema the cellular construction site that salamanders use to regrow limbs), and the second told that construction site what to put up. Out came bone, joint, tendon and ligament. Not a flawless replica, but unmistakably the right parts in roughly the right places. 

🧐 What's in it for me? Regrowing a finger on demand remains a way off, and the mouse results were partial (sorry mouse), but in the short term, nudging wounds away from scarring and towards rebuilding via the same process is extremely tangible. Better healing first, spare parts later.

💵 Out of the Lab: If scarring is a choice the body makes rather than a limit it's stuck with, the regenerative medicine pitch shifts from importing cells to reprogramming the ones already present, which is cheaper, simpler and a great deal more scalable.

Artificial photosynthesis has a tedious problem. Plants make it look effortless, turning sunlight, water and CO2 into stored energy without so much as a login screen, whereas our versions need batteries and a tangle of electronics just to cope with clouds drifting past. 

Sunlight is inconveniently variable, and wringing steady power from it usually means an expensive box of chemicals to smooth everything out. However, researchers have now done away with the box. They built a solid electrolyte into the device so that, as sunlight intensifies and the electrolyzer warms, its electrical resistance obligingly drops and current flows more freely. The thing tunes itself, no batteries required, and outdoors it kept producing formic acid (a tidy liquid that stores energy and can release hydrogen later) even as the sun came and went.

🧐 What's in it for me? Stripping out the batteries and electronics is the unglamorous part that matters a surprising amount, because cheap and simple is what lets a lab demo become a thing on every house.

💵 Out of the Lab: The quiet thesis here is that the future of solar might not be electrons on a wire but molecules in a tank, with formic acid as a cheap, stable way to carry hydrogen around without the drama of compressed gas.

SETI is an institute that searches for extra-terrestrial life. For more than sixty years they’ve been listening for a very particular kind of alien, namely, one polite enough to broadcast a razor-thin radio signal, the sort nature doesn't produce by accident. 

But a new paper suggests we may have been waiting for the wrong shape. Before an alien signal ever leaves home, it presumably has to barge through the stellar tantrums of its own star, which smear a sharp spike into a broad, faint smudge (exactly the sort of thing our detectors are built to ignore…) Worse, the effect is most pronounced around M-dwarf stars, which make up roughly three-quarters of the galaxy. 

Luckily, the fix could be straightforward (just widen the search), which raises the mildly deflating possibility that ET has been calling for decades, we just haven’t picked up. 

Until next time.

Like what you're reading? Share toast with a friend.