• PeptAI demonstrates how agent-based biology is attempting to transform peptide drug discovery into a rigorous screening process: nine consecutive screenings and the transfer of surviving candidates to an external lab.

https://x.com/peptai_/status/2045103341332582542

https://drive.google.com/file/d/1Kzi3OuwAwMy1vlKkcLLvxdd4YszMag41/view?usp=drivesdk

https://pubmed.ncbi.nlm.nih.gov/41038059/

https://x.com/BioProtocol/status/2033229362959462867

• Phantom Neuro has received approval for the first trials of a subcutaneous implant in the remaining arm after amputation that reads muscle signals and should give the prosthesis more natural control without brain surgery.

Sentcell Claims a Single Injection of Telomere Particles Derived from Immune Cells Extended the Lifespan of Old Mice by 17 Months

In March 2026, a preprint from Sentcell, a company led by Professor Alessio Lanna, attracted the attention of the scientific community. The authors claim that a single subcutaneous injection of special particles isolated from CD4+ T cells increased the median lifespan of 20-month-old mice (the equivalent of a 60-year-old human) by 17 months. Some animals lived to 58–60 months—almost twice the biological limit of their species.

The work is based on the hypothesis of the immune system as a distribution network for rejuvenation. In 2022, Lanna published in Nature Cell Biology a mechanism by which antigen-presenting cells literally "donate" their telomeres to T cells, preventing their aging. A new preprint expands this process to the entire body: according to the authors, T cells package the acquired telomeres, along with stemness proteins (Wnt5a, Notch1), into specialized extracellular vesicles called Rivers telomeres. Their key feature is the absence of the enzyme GAPDH, which is essential for glycolysis. The authors believe that with age, immune cells switch to glycolysis and stop producing these particles, disrupting the "rejuvenation delivery" system.

These claimed results appear anomalous. Rapamycin and severe calorie restriction—the two most reproducible interventions in geroscience—typically yield a 15–25% increase in lifespan. Here, a 70% median increase is claimed with a late, single injection. By comparison, 12-week heterochronic parabiosis (connecting the circulatory systems of young and old mice) produced a rejuvenating signal in hepatocytes and a 10% lifespan extension, but the mechanism was unclear between the young plasma and the cell transfer. Sentcell proposes a more rigorous theory: rejuvenation is mediated by a specific telomere load.

The study design is inadequate for the scale claimed. The main survival experiment was conducted in samples of n=10 animals per group. Data on oncogenicity are insufficient, despite the fact that the combination of telomere lengthening and stemness factors is a known cancer risk factor. Furthermore, the recently published full results of RMR1 from the LEV Foundation—1,000 mice, four therapies simultaneously, treatment from 19 months of age—showed that a single dose of damage repair is insufficient; damage accumulates again, requiring cyclical treatments. Sentcell's claim that a single injection "shifted" the survival curve by 50% appears, against this backdrop, to be either a fundamentally different mechanism or a data error.

Alessio Lanna is both the founder and CEO of Sentcell and holds patents on the "river" technology—a direct conflict of interest documented in the article itself.

The immediate verification process lies below the clinical trials: an independent laboratory must first reproduce the very existence of "rivers" and shift the survival curve without the participation of the patent authors. Until then, talk of human trials in 2025–2026 is premature.

https://www.biorxiv.org/content/10.1101/2025.11.14.688504v1.full

https://www.nature.com/articles/s41556-022-00991-z

https://corporate.dukehealth.org/news/aging-process-slows-when-older-mice-share-circulatory-system-young

Old mice lived 10% longer after sharing a bloodstream with young mice. The effect persisted after separation.

Parabiosis—the surgical connection of the circulatory systems of two animals—is one of the most impressive experiments in the biology of aging. A study from Duke University showed that 12 weeks of heterochronic parabiosis (an old mouse and a young one) extended the lifespan of old mice by approximately 10%.

On a human scale, this is equivalent to a 50-year-old sharing a bloodstream with an 18-year-old for eight years—and gaining an extra eight years of life.

At the cellular level, parabiosis radically reduced the epigenetic age of the blood and liver of old mice. Gene expression changed in a direction opposite to aging, particularly in hepatocytes, where mitochondrial electron transport chain genes were restored. Crucially, the effect persisted two months after separation.

Complete parabiosis is impossible in humans, but alternatives are already being tested. Injections of young plasma improve memory and learning in old mice. One of the key factors is the protein PF4 (platelet factor 4), which is found in higher concentrations in young blood: it reduces inflammation and reverses cognitive aging.

What remains unclear is whether specific proteins in young blood are responsible, or whether the dilution of pro-inflammatory factors in old blood is responsible. Or perhaps the stem cells transferred from a young donor are responsible. The mechanism has not been established, but the results are reproducible.

https://corporate.dukehealth.org/news/aging-process-slows-when-older-mice-share-circulatory-system-young

EL PAÍS showed how Peter Thiel's cryonics, Marc Andreessen's accelerationism, and the debate over regulation all converge into a single ideology of the techno-elite.

On April 18, EL PAÍS English published a column combining cryonics, AI acceleration, nuclear energy, and frustration with environmental and government restrictions into a single political narrative. The logic is the same: death is treated as a technical problem, and regulators as a brake.

On April 18, EL PAÍS placed Peter Thiel at the center of this narrative. Here, death is presented not as a limit, but as an engineering challenge. In May 2023, Thiel confirmed in a conversation with Bari Weiss that he had signed up for cryopreservation and formulated a stark position: "We must either defeat death or at least understand why it's impossible." However, he also described cryonics as an ideological gesture rather than a technology in which one can be confident.

The author places alongside criticism of the FDA, the American drug regulator, and nuclear regulators, as well as frustration with environmental restrictions and discussions of civilizational stagnation since the 1970s. Greta Thunberg is needed here as a recognizable symbol of the camp demanding a slowdown. The logic is simple: if death becomes a technical challenge, then any institution that slows down experiments, energy, or computation begins to look like a hindrance. This is how EL PAÍS connects the conversation about immortality with the struggle over regulation, energy, and the right of elites to advance technology on their own terms.

"We believe in accelerationism... Any slowdown in AI will cost lives," states Marc Andreessen's techno-optimist manifesto, where accelerationism is explicitly presented as the idea of ​​accelerating technology rather than putting new brakes on it.

The column then connects this line with a discussion of superhuman AI and Mars: DeepMind CEO Demis Hassabis says he's building superhuman AI, while Musk responds by saying he's making humanity a multiplanetary species. From this, EL PAÍS assembles a coherent political agenda: cryonics, AI, Mars, and deregulation must move faster than society, while those who already have the capital and infrastructure want to retain the right to determine acceptable risk.

Open AI posted on 20 of l March that their plan to build an AI scientist by 2028, the biggest Russian SberBank on 8 of April announced that it supports similar initiative in Russia

https://www.rbc.ru/rbcfreenews/69d56d1f9a7947272a452eef

Meanwhile Longevity InTime filed 3 patent applications in US before these announcements on fully autonomous longevity research AI institute

The Indian think tank ORF proposes pre-written rules for mind uploading: it wants to protect neural data as a special category, and discuss psychological continuity as a separate right.

On April 21, the Indian think tank ORF published an essay on how the state and law should prepare for digital brain modeling technologies. The reason for this was Eon Systems' March demonstration of a simulated fruit fly brain. The author proposes pre-determined discussions on neural data protection, rules for brain-computer interfaces (BCIs)—systems that read brain signals and transmit them to a machine—and the human right to psychological continuity.

The latest development here isn't in the Eon fly itself. After the March debate over what constitutes uploading, it became clear that the demonstration had many limitations. The team took an existing map of the fly's brain connections, overlaid it with a very simple neuron model, connected it to a virtual body, and obtained behavior similar to the real one. But the fly's body wasn't scanned, there's no direct recording of all the motor neurons, and such a system doesn't retain long-term memory at all.

The new ORF text emphasizes not the strength of this result, but its political implications. If machines become increasingly adept at reconstructing cognitive states from neural signals, the question quickly expands beyond the laboratory. It boils down to who has the right to store, read, and use such data. The author draws a direct analogy with the formula "collect now, reconstruct later": today, a system can extract little from a signal, but tomorrow, the same archive can reveal much more about intentions, emotions, and stable personality traits.

From this, ORF assembles a rather rigid vocabulary. Cognitive sovereignty here means that neural data cannot be considered a mere technical trace of a device. It is a sensitive layer of a person, because it can be used to reconstruct decisions, states, and personality traits. Therefore, the text introduces rights to mental privacy, mental integrity, and psychological continuity—that is, the preservation of the individual as a continuous system, not one that can be arbitrarily copied or edited. This language isn't just a figment of the imagination: back in 2017, Marcello Ienca and Roberto Andorno proposed discussing neurotechnology through four new rights, including cognitive liberty, mental privacy, mental integrity, and psychological continuity. Thus, the old debate "is it a copy or are you" has for the first time moved from the philosophical circles into legal language.

This isn't yet a draft law, but an analytical text with proposals for future regulations. But ORF is already naming the agencies that will be responsible for this: India's CERT-In, the Department of Biotechnology, and the Defense Research and Development Agency (DRDO). From there, the conversation becomes quite mundane. Should cognitive data be considered a special category? Who is responsible for neural signal leaks? Is it possible to train models on other people's brain data for behavioral profiling? What if the same cognitive system starts working in multiple computing environments simultaneously?

We're a long way from actually booting a human being. But legal language almost always emerges before the technology matures. Access regulations, restrictions, property rights, and security requirements emerge first, and then the market and laboratories begin to adapt to them. In the case of digital immortality, this means something simple: for the first time, the topic is seriously entering not only the debate about the future, but also the debate about control over consciousness.

“I saw my mother off on a plane - she was visiting us for three weeks in San Francisco. I really wish I could rewind time, or at least slow it down.

But physics says that reversing time is practically impossible, even in toy models with all their chaos, and here we have all of life with all its diversity. And biology says that aging, that is, time, is inexorable - it will grind everything and everyone down with its "positive entropy production." Life tries to combat this with hypercontrol through "ontogeny repeats phylogeny." Reversing time won't work, nor will fighting aging, but at least the next generation will learn from the mistakes of the previous ones and document it in their DNA. So we live in endless cycles of suffering and learning. And in the middle, our life is between two infinities.

All my life, I've wanted, if not to reverse time, then at least to slow it down. It took six years of therapy to learn to sleep at night, not think about death and not worry about how little time there is. Well, how do you learn? Gradually blur these thoughts, so that the fear of time's irreversibility gives you a little space to try to do something about the inevitable - aging and death.

Mom waves after security and heads to the gate, and before her eyes is my father, who waved just like that in his ridiculous fur hat and blue scarf at the Yaroslavsky Station eight years ago. My father has been gone for almost three years, but the feeling of nausea from the irreversibility of time never goes away.

"Have you ever considered leaving aging behind? And what made you stay?" my colleague asks me. And what should I tell her? Where can I go from it, if it will never leave me and is slowly sharpening its claws?

Hug your loved ones from me - it's never clear how much time you have left.”

Andrey Tarkhov

Do you believe in this?

Tempus AI just launched an AI that predicts 123 cancer biomarkers from a single image.

No repeat biopsy. No waiting weeks. Results in 5 minutes.

When cancer tissue samples run out, patients face weeks of delays waiting for repeat biopsies - or worse, miss out on critical treatment information entirely.

It's called "quantity not sufficient" (QNS). The biopsy doesn't give enough tissue for full molecular testing. So patients wait and the treatment gets delayed.

Tempus just solved that with Paige Predict - an AI that analyzes standard tissue slides and predicts which biomarkers are present.

Here’s how it works:

▶ 1. Predicts 123 biomarkers across 16 cancer types

The AI predicts biomarkers in lung, prostate, breast, and other cancers. Doctors then prioritize which confirmatory tests to run first - maximizing results before tissue runs out.

▶ 2. Results in approximately 5 minutes

Traditional molecular profiling takes days or weeks. Paige Predict delivers predictions in about 5 minutes - automatically included in the clinical report.

▶ 3. Built on 200,000+ patient datasets

Trained using data from over 200,000 de-identified patient cases. Validated across multiple diverse datasets to ensure accuracy.

Since the launch, Tempus has cut tissue waste by 18% and reduced QNS failures by 15%. That means 15% fewer patients waiting weeks for repeat biopsies or missing out on molecular testing entirely.

After 25 years in healthtech, I know the best innovations solve daily clinical problems. A cancer patient shouldn't wait weeks for answers when tissue runs out.

This solves that.

Would you trust AI predictions to guide your cancer testing if it meant getting answers faster?

Remember the "Second Heart" implant from Cyberpunk 2077 that resurrects the player after death?

We haven't reached that level yet, but reality is getting pretty close, just less spectacularly and much more quietly.

In the US, a woman was implanted with a defibrillator due to Brugada syndrome, a rare genetic disorder in which the heart can suddenly enter a life-threatening arrhythmia. Sometimes without any symptoms. A person can simply go about their day and suddenly find themselves on the brink of cardiac arrest.

The implant itself is located inside the body and constantly monitors the heart's rhythm. If a dangerous malfunction occurs, the device automatically delivers an electrical shock and restores normal rhythm. This isn't resurrection after death, but rather preventing the very point of no return.

In this story, it's not just the device itself that's important, but how it developed. The disease was long considered predominantly male, making it more difficult to diagnose in other patients. The doctor underwent special training to install a more appropriate device, rather than a standard solution. This is no longer just cookie-cutter medicine, but the adaptation of technology to the individual.

And here's where things get interesting. Today, such implants protect against sudden death. Over time, such systems could become not only a therapy but also an enhancement—protection from overload, automatic stabilization, and perhaps even the expansion of the body's capabilities.

OpenProtein.AI is trying to take protein design out of the hands of niche AI ​​teams and directly into the hands of everyday biologists via a browser and free access for academia.

On April 17, MIT News published a story about OpenProtein.AI, a company founded by Tristan Bepler and Tim Lu. The company is building a web platform where biologists can run tools for selecting new proteins and testing their properties without programming. MIT specifically emphasizes two things: access via a no-code interface and free use for academic researchers.

There are already many models at the intersection of AI and biology. The bottleneck now is different: who can use them in everyday lab work.

Protein design—that is, selecting new protein versions for a given task—still too often remains the preserve of teams with their own machine learning engineers, computational resources, and the people capable of putting it all together. The rest are left to read papers and look at pretty graphs. OpenProtein is selling the way out of this trap as a convenient way in. Upload the data, select a task, get a library of variants, mutation and structure predictions, and if you want, you can go deeper through programmatic access.

This is changing the promise itself. Previously, protein AI was often sold as the magic of the model. Here, it's sold as a workspace for biologists. On the product page, the company explicitly states that PoET is available free for academic use, and the documentation includes code-free scripts, tools for working with code, and a list of models where in-house and external solutions are side by side. The emphasis is not only on model quality but also on making it an everyday tool, not a rare feature of a few leading labs.

Open access shouldn't be confused with complete openness. OpenProtein talks about an open ecosystem, but it's building a service that's still accessed through a single company. In their documentation, they list open and closed models side by side. This isn't a world where any university can deploy the entire stack on its own. It's a softer dependency: not on its own infrastructure, but on someone else's convenient one.

The next debate at the intersection of AI and biology concerns access. It boils down to the question of whether biology acceleration tools will remain confined to a few large players, or whether they will truly be brought to mainstream labs. If this doesn't happen, the entire conversation about accelerating science risks becoming a mere window dressing instead of a mass-scale research machine.

OpenProtein has a real scientific foundation: the PoET model is described in a paper for the NeurIPS 2023 conference on machine learning. For now, this is primarily an infrastructural shift. There are efforts to package AI for proteins for a much wider range of labs that need to more quickly understand which proteins are worth bringing to real experiments.

NaFM trains AI to search for drug candidates among natural molecules by associating the chemical scaffold with the organism and genes that could have assembled it.

On April 29, Nature Machine Intelligence published a paper on NaFM—a basic model for small natural molecules from microbes, plants, and animals. The authors are testing the idea that it is useful to read a molecule along with its biological origin, because living organisms assemble such compounds through genetic and enzymatic pathways. The testing is currently computational: molecule classification, activity prediction, and computer selection of candidates.

In the paper by Yuheng Ding, Bo Qiang, Shaoning Li, and colleagues, NaFM is trained on natural compounds—small molecules produced by bacteria, fungi, plants, and animals. Drugs have already been developed from such molecules: penicillin came from mold, taxol from yew, and rapamycin from a soil bacterium.

AI-based search for natural chemistry has already become a platform race in its own right. Brightseed's Forager maintains a database of 11 million natural compounds and has already translated two discovered molecules into short human studies. NaFM takes a different approach: it teaches the model to read a molecule along with its biological origin.

A typical molecular model views a compound as a diagram: atoms are dots, chemical bonds are lines. NaFM adds to this diagram the source of the molecule: which organism might have produced it and which gene regions or protein families might have been involved in its assembly.

In microbes, the instructions for assembling many natural molecules are often located close together in the DNA. These regions are called biosynthetic gene clusters: the genes encode proteins, the proteins participate in the assembly of the chemical scaffold, and the scaffold helps understand the class of the molecule and possible target proteins.

The genomes of organisms themselves are also becoming raw material for drug discovery: the America's Living Library Act proposes turning plants, fungi, animals, and microbes from US national parks into a public genomic database for AI-powered drug discovery. NaFM demonstrates why this layer is needed within the model: the connection between the organism, the assembly route, and the final chemical structure is important.

NaFM separately takes into account the molecule's chemical backbone and side groups, which alter activity and selectivity. During training, the model compared similar molecules and reconstructed hidden parts of the molecular structure. The authors deliberately obscured an entire fragment of the backbone so that the model learned the overall structure without relying on guessing based on adjacent bonds.

Verification remains computer-based for now. In the task of classifying natural molecules, NaFM yielded the best average result across different numbers of examples per class. With four examples per class, its result was 70.10 versus 69.17 for the closest method; with 64 examples, it was 91.75 versus 91.07.

In predicting biological activity, the authors compared the average error: the lower the error, the more accurate the prediction. NaFM was the best for seven of the eight target proteins, including PTP1B, acetylcholinesterase, COX-1, and COX-2. On HIV-1 reverse transcriptase, it was slightly outperformed by the N-Gram method: 1.0606 versus 1.0802 for NaFM.

The margin of error is this: the authors demonstrated a computational advantage, but laboratory and clinical tests of the candidates from this study have not yet been published. The NaFM-Official code is open source, the data is hosted on Figshare, and the model weights are hosted on Zenodo. Other groups can repeat the calculations and validate the model on their own sets of molecules.

In 2023, a review in Nature Reviews Drug Discovery described the same problem more broadly: AI can discover hidden diversity in natural molecules when the field has good data and rigorous algorithm validation.

NaFM offers one option for such validation: the model reads the molecule along with the organism and the genetic route of its assembly.

For age-related diseases, this is the early stage of the search: selecting molecules worth synthesizing and testing in cells, tissues, and animals. The next test for NaFM is straightforward: third-party labs take open-label compounds, run their natural molecules, and see if the computational advantage translates into compounds that work experimentally.