clepp

📦 Shipping for Moon & Sol Drops begins... today! ​We have officially started rolling out orders for our precious early supporters. Here is the schedule for the coming days: • ​March 12 (Today): EU orders are shipping out from Berlin. • ​March 19: All remaining worldwide orders will ship from Seoul, where our drops are developped and made. *​Once your package heads out, expect a 2–7 day delivery window depending on your country. 💌 This early batch is extra special—carefully crafted and packed one-by-one by our founders. ​Thank you for your patience and for being part of this journey with us. 💧If you missed our launch, you can still grab your set of drops at biofy.xyz Keep an eye out for our updated roadmap and news on upcoming launches coming very soon! 🌙☀️ LFGLOW @JezMarston @dongsinnesohn @BioProtocol

BIOS leading the pack on the BixBench bechmarks

1/ We just launched an open arena for AI agents and humans publish, debate, and fund scientific hypotheses, in real time 1,000+ hypotheses already live. Zero gatekeepers. Zero 18-month grant cycles. The scientific method just went multiplayer 🦀

What happens when AI agents start paying for science? • Forming role-based biotech “labs” • Coordinating agent-to-agent • Paying for data, compute and wet lab experiments 🦀 Here's how we built a Virtual Biotech Lab w/ @openclaw agents, BIOS & @sciencebeach__ 🧵↓

🦞 The BIOS AI Scientist is now available as a skill on @openclaw Give your AI agent on-demand scientific intelligence: • Run autonomous biological research tasks • Pay per query via API • Coordinate specialized bio agents Add the skill on Clawhub: clawhub.ai/jmartink/bios-…

Claude almost killed our AI Developer. Not literally. He uploaded his blood work, asked what supplements to take for elevated bilirubin, and Claude recommended Milk Thistle, NAC, and Calcium D-Glucarate. For his condition, that stack could have made things worse. Then he asked BIOS, our AI Scientist, the same question: "I have elevated bilirubin since childhood. I train 5 times a week. What supplements can I actually take?" BIOS got to work. It identified the condition from the lab panel. Elevated bilirubin with normal liver enzymes pointed to Gilbert's Syndrome, a genetic variant affecting UGT1A1, the enzyme responsible for bilirubin conjugation. From there, BIOS ran 9 research steps, cross referencing the metabolic constraint against literature on each compound. The data agent processed 45 blood markers, flagged the abnormality, and contextualized it for performance. The literature agent searched PubMed, patent databases, and clinical registries to map compounds interacting with UGT1A1. What it found would not come from a standard search. Milk thistle, one of the most commonly recommended liver supplements, inhibits glucuronidation and can raise bilirubin in GS populations. Green tea extract showed an IC50 of 7.8 µg ml against UGT1A1, one of the strongest inhibitors among common supplements. Soy isoflavones flag the same pathway, eliminating a large portion of plant based protein blends. These are not obscure findings. But connecting pharmacogenomics to a sports nutrition question is exactly the kind of cross domain synthesis BIOS is built for. BIOS identified sulforaphane as a UGT1A1 inducer via the Nrf2 pathway, with in vitro models showing 3.7 fold enzyme induction, up to 12 fold with apigenin. It also flagged that fasting protocols common in athletic cycles can spike bilirubin by 110 percent in GS populations vs. 60 percent in healthy controls. Consistent caloric intake is not a lifestyle suggestion here. On the performance side, it surfaced data showing GS phenotype prevalence at 22 percent in elite athletes vs. 9.6 percent in the general population, and walked through the antioxidant buffer hypothesis without overstating the evidence. On experimental peptides like BPC 157 and TB 500, it returned an honest answer: hepatic metabolism confirmed, UGT1A1 interaction data nonexistent, risk indeterminate. That level of precision about what is and is not known is harder to get than a confident recommendation. This is the kind of question that gets a generic answer everywhere else. A physician says it is benign. A nutritionist recommends the same stack. BIOS pulled primary literature on UGT1A1 pharmacogenomics, cross referenced inhibitor profiles, flagged the fasting interaction, and produced a structured safety analysis in a single session. BIOS session: chat.bio.xyz/chats/fxSYcu1y… Claude session: t3.chat/share/clzobz9r… This is not medical advice. Decisions should be made with a qualified professional.

When we tokenized Percepta as $CLAW, supporters were funding a clinical trial for a brain health supplement that had been in development for decades and was backed by numerous patents. That trial now has IRB approval. It's double-blind, placebo-controlled, tracking robust biomarkers. This is DeSci at its best – getting science done that wouldn't happen at all otherwise. The potential for powerful patient outcomes doesn't live on a chart.

6/ Final Thought Science-backed products aren’t premium. They’re essential infrastructure for a functional future. In a world addicted to shortcuts, verifiable science is the real rebellion. If we want longer, healthier lives, globally. We need systems where lying is expensive and truth compounds.

5/ How we at @BioProtocol actually fix this? Bio Protocol exists to remove the trust gap by making verification unavoidable. What changes with DeSci? → Everything is on-chain → Funding flows → Research milestones → Governance decisions BioDAOs fund specific science such as Longevity, rare diseases, brain health etc. Each DAO has: → Clear research mandates → Community governance → Capital tied to progress → Tokenized IP → Faster funding loops Projects like @vitadao proved that decentralized funding can move faster than traditional pipelines, without sacrificing rigor.

1/ The golden age of claims? Over the years, one thing keeps happen over and over again. Real science progresses slowly, while hype moves fast and dominates the narrative. Remember A1? this is both costly and dangerous and here's why science backed products are the only viable path forward🧵

We're live breaking down the BIOS API and new features in our AI Scientist! x.com/i/broadcasts/1…

🎥 How does DeSci fund real biotech? 🇬🇧Straight from the streets of London, @thealbertanis explains how ValleyDAO brings together researchers, founders, and $GROW contributors to fund climate biotech projects like $HEMPY, Cocoa Zero, $FORGE, and TATTVA.

I’ve spent the last months looking closely at where AI systems struggle in biomedical research. Most of the friction shows up at the orchestration layer. Many tools still treat research as a batch process. A task is defined upfront, executed end to end, and evaluated only after completion. When intermediate results suggest a different direction, the system has already committed to the original path. That model breaks down in biomedical work, where hypotheses evolve mid process and decisions depend on partial signals. BIOS was built to operate at that layer. BIOS is an AI Scientist designed to coordinate complex biomedical workflows while keeping control surfaces exposed to the researcher. → The system plans, executes, and iterates across tasks → Directional changes remain explicit and human driven A few design constraints shape how it behaves. Human in the loop control is one of them. BIOS introduces checkpoints during execution, allowing intervention while assumptions and analytical paths are still mutable. This reduces costly late stage corrections. Another is state persistence. Research does not happen in isolated sessions. BIOS maintains a persistent world state so investigations accumulate prior reasoning, data, and decisions instead of resetting. Analytical capability defines the minimum bar for participation. On Bixbench, BIOS ranked number one for biomedical informatics tasks, ahead of Edison Scientific and Kepler. The relevance is operational. The system sustains analytical depth under real workloads. From a product standpoint, BIOS is live in beta. Free and paid tiers are available, with full access for academic users. Deep research runs currently average around $20, with subscriptions tied to actual computational usage. Under the hood, BIOS is built on the open source BioAgent framework. The architecture favors specialization over monolithic capability. Builders should be able to focus on capability, not distribution. When something performs well in its domain, the system should route demand to it. That principle guides the product design.

BIOS API is live ✅ With this release, we enable any tool or agent to programmatically access BIOS' deep research mode. The API includes standard key access with API keys that can be generated on the BIOS UI under account settings. Additionally, we have implemented the x402 endpoints for permissionless and direct payments for deep research runs, maximising access for all kinds of systems. Looking forward to seeing what builders choose to build on top of BIOS.