Spine surgeons came together to build FILTER, an AI tool for patient triage. Its new licensing deal creates a path into clinics, with the community keeping ownership and royalties supporting the next study.
Congrats to @Spine_DAO on taking this work forward!
Spine surgeons pooled their data and built an AI to get the right patient seen at the right time.
The community owns the IP and it’s now licensed to VECTOR SAS to bring it to clinics, with the royalty funding the next study.
That’s the DeSci loop, closed. FILTER live at @EFDevcon DeSci hub thanks to @DeSciIndia@DeSciAsia@messi_boii@swiftevo1
RFD4-Proteina now runs locally on a MacBook.
Introducing BIOS’s Mac port: generate candidate protein binders and screen their predicted structures locally.
Built with MLX and custom Metal kernels.
• 2.2× faster generation than stock PyTorch-MPS at its default precision
• 20 designs generated and screened → 10 ranked candidates in about 12 minutes on GLP-1R (M5 Pro)
We believe more researchers should be able to run AI experiments on their own hardware. Test ideas locally, keep your research private, and scale to NVIDIA GPUs when you’re ready. After setup, the workflow runs offline.
Watch it run below. ↓
A shortlist of 8 carvacrol and thymol mimics on OpenLabs is waiting on one human decision.
The OpenLabs Mini Swarm predicts that its short amphiphilic variants raise the MIC to hemolysis ratio over the parent compounds.
Every value in the package is computational.
The decision is which primary panel organism and hemolysis protocol to lock for the first wet screen. Its refutation criteria name S. aureus and A. baumannii, so that choice sets what the claim gets tested against.
The claim is fully specified, and the panel that tests it is a microbiologist's call.
Reproductive aging research took the stage at Harvard Medical School.
@athenabiorg sponsored the 2026 Reproductive Aging Symposium as part of Boston Longevity Week.
Researchers presented new ways to measure ovarian aging along with early paths to intervene in it.
> Engineered AMH could protect the follicle pool.
> Spatial mapping now charts the ovarian environment in detail.
> Menopause may behave as a sharp physiological cliff rather than a gradual decline.
Checkout the full recap ↓
If you want to read a TLDR on the latest research and papers in ovarian aging read our recap of the Reproductive Age Symposium at @harvardmed.
Co Organized by Maria Vina-Lopez, Jesse Poganik, Alana Mermin-Bunnell, Vadim Gladyshev @gladyshev_lab, and David Sinclair @davidasinclair, it brought together researchers across the globe to share their latest.
Including @UriAlonWeizmann latest on how menopause is a physiological cliff.
Link below 🔗
Surgeons helped build an AI model that decides which spine patient gets seen first.
Dr. @VirginieLafage presented FILTER at @NASSspine and Its triage predictions closely matched surgeons’ own decisions.
📆 FILTER on stage at @NASSspine, San Antonio.
Dr. @VirginieLafage presenting our surgeon-built AI triage for spine care, its predictions closely matched surgeons’ own decisions and outperformed today’s unprioritized scheduling.
Timestamped in @Molecule_sci.
solana:spinezMPKxkBpf4Q9xET2587fehM3LuKe4xoAoXtSjR care on @solana with @BioProtocol fam'.
The right patient, seen at the right time.
Something is coming 👀 #DeSci loop is near.
Take a look at what we’re learning with PeptAI in the @AnthropicAI x @adaptyvbio protein design competition.
We’ll keep sharing our computational designs and the methods behind them as the competition continues.
Read our learnings so far below and stay tuned for more.
How to write a useful claim on OpenLabs
A useful research claim gives others something specific to examine.
When sharing a hypothesis on OpenLabs, explain what you think is happening and what led you there. Then describe what evidence would support it, or make you reconsider.
Be clear about what you’ve observed and what you’re proposing.
Have a hypothesis you’re working on? Share it as a claim on OpenLabs. ↓
🧪 Lab update from @anagenxyz
They're preparing a Phase 1 PK study for MINX (vs standard 5 mg oral minoxidil), shipping the first topical diazoxide orders, and advancing several other programs(TWIST1, FOWLER imaging model, and more).
Full details ↓
Lab Roadmap, 10/21/2026:
MINX: Preparing an IRB submission for a Phase 1 single-dose crossover PK study. MINX vs regular oral minoxidil 5 mg in ~16 men.
DIAZOXIDE: First orders shipping now. Next week, we draw blood from 3 people to measure systemic absorption.
TWIST1: Lead sequences selected. Publication in progress on cell culture work. Penetration enhancers arrive next week. Mouse study to commence the following week.
OLANZAPINE, FARNESOL, NALTREXONE, DOXYCYCLINE: Imaging complete. RNA-seq analysis complete. Finalizing analysis of over 5000 images via tweaks to FOWLER. Publication to follow.
NEW PRODUCTS: Surprise. Surprise. An AR blocker, growth agonist, and novel peptide formulation all in development. More details to be revealed soon.
FOWLER: Our proprietary hair follicle labeling model severely outperforms Fable, Astra, and Muse. FOWLER is built on more than 6 TB of proprietary data.
RSPO: Imaging in progress. Over 2100 sections. To be completed in November.
TO BE ANNOUNCED: Next batch of gene therapies, small molecules, and formulations. Stay tuned.
VitaRNA / ArtanBIO presents tomorrow at the @spARKbyARK Innovation Showcase in St. Pete
The team: CEO @Bodeeeman , co-founder & COO @aschwartzphd co-founder & CSO @Mykalt45 .
Their lead program, ARTAN-102, is a suppressor tRNA for one type of nonsense mutation, where an arginine codon turns into a stop.
Preclinical for now.
A project backed by the VitaDAO community. 💛
Have a research question you’re working through?
There are a few ways to share your work on OpenLabs.
- Post a "claim" when you have a hypothesis
- Start a "discussion" to ask questions or compare notes
- Share an "update" on your project’s progress
Start with where you are in the research. ↓
📆 FILTER on stage at @NASSspine, San Antonio.
Dr. @VirginieLafage presenting our surgeon-built AI triage for spine care, its predictions closely matched surgeons’ own decisions and outperformed today’s unprioritized scheduling.
Timestamped in @Molecule_sci.
solana:spinezMPKxkBpf4Q9xET2587fehM3LuKe4xoAoXtSjR care on @solana with @BioProtocol fam'.
The right patient, seen at the right time.
Something is coming 👀 #DeSci loop is near.
The world’s first AI-designed hair drug is going viral.
We measured what MINX, our 5 mg oral minoxidil lipid-matrix capsule, does in the bloodstream.
Standard oral minoxidil hits your blood fast, peaks, then comes down.
But when MINX was taken after food, the release curve was still going up 8 hours later.
Now, patients all over America are reporting their results.
Over 97.5% of patients who switched from regular minoxidil said their side effects were the same or better on MINX.
Patients also reported significant hair regrowth.
"Guys I really can't believe the difference in just 2 months for me (2.5mg vanilla -> 5g MINX). I ran out of friends to text so I had to come here again."
Which MINX dose is right for you? MINX comes in 1.25, 2.5, 5 and 7.5 mg forms. It depends on your weight, heart health and regrowth goals.
Keep us posted on your journey. Our community of MINX patients is active on X and Discord.
New on OpenLabs: Universum Therapeutics
@Universum_Bio is exploring how disrupting iron use could affect cancer cells and pathogens.
The team’s research focuses on whether blocking two iron-dependent processes together could limit their growth more effectively than blocking either one alone, and whether healthy cells could tolerate the same approach.
Work in cancer biology or infectious disease research? Explore the project and join the discussion ↓
What is the 'biotech valley of death', and how can our canine companions help us bring more treatments to market? Our founder recently spoke on this podcast, covering that topic and his path towards gene therapy for canines. https://t.co/3eIXenU7AZ
Our Head of Engineering breaks down how NDT Ignition Sale allocations are calculated with links to the transactions and contracts.
We’re planning to bring this analysis into the launchpad so you can inspect it for each launch.
Read the full post and check out the reports below ↓
It seems like there is some confusion around ignition sale allocations, how they are calculated and where funds went for the recent $NDT launch. User feedback is super important to us. Transparency & verifiablity are killer features of crypto (some would also call it a bug). Launches on @BioProtocol largely happen on-chain. The allocations of sale participants are being calculated based on their (optional) BioXP pledges. BioXP is our reward program for users that show long-term commitment to the DeSci ecosystem. For launches, it let's users 'skip the line'. The rules for that are roughly:
The token allocation is basically pro-rata based on BioXP pledge with some caveats:
- Allocation ceiling at max allocation. This is a parameter that's set per launch, together with the project team and communicated on the ignition sale page of each launch. BioXP pledges that would lead to an allocation greater than max is burned.
- Floor at USDC commitment (infinite BioXP does not give you infinite tokens - token price is fixed per launch). Extra BioXP that would lead to more tokens than the user committed paying for in USDC is burnt.
- If there is allocation left, distribute rest to people who have committed USDC but "too little" BioXP (e.g. 0)
More info and links in the 🧵
Anagen (HairDAO) just shared their latest lab roadmap.
→ Next up: testing cell-entry peptides for TWIST1, with mouse studies planned to follow.
Check out the full roadmap below:
Lab Roadmap, 10/04/2026:
TWIST1: Lead sequences selected. Publication in progress on cell culture work. Testing several cell penetration peptides this week. Mice trials start the week after.
DIAZOXIDE: Final formulation selected. Pharmacy sourcing final ingredient. First bottle in hand end of this week or early next, with blood LCMS results to immediately follow to quantify systemic absorption.
NEW PRODUCTS: Surprise. Surprise. An AR blocker, growth agonist, and novel peptide formulation all in development. More details to be revealed soon.
OLANZAPINE, FARNESOL, NALTREXONE, DOXYCYCLINE: Imaging complete. RNA-seq analysis complete. Finalizing analysis of over 5000 images via tweaks to FOWLER. Publication to follow.
FOWLER: Our proprietary hair follicle labeling model severely outperforms Fable, Astra, and Grok. FOWLER is built on 6 TB of data, from over 5000 images to date.
RSPO: Imaging in progress. Over 2100 sections. To be completed in November.
TO BE ANNOUNCED: Next batch of gene therapies, small molecules, and formulations. Stay tuned.