Bio Protocol

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__ 🧵↓

Gates are opening soon.

Bio Protocol designed a novel ADHD drug candidate in 24 hours using AI. First lab test: $500. Full validation: ~$1,500. Pharma spends millions and years to reach the same point. That changes what DeSci can do. Drug development has always been gated by capital, not science. Research gets funded when it aligns with commercial priorities. Plenty of valuable questions never get asked because no one with the budget chose to ask them. DeSci has been working on that gap for years, but the ambition stayed kneecapped by the cost of discovery itself. AI changes the math. With costs this low, a DAO treasury can fund a drug candidate all the way validation with real lab data. Entire categories of research pharma was NEVER going to touch are now within reach. Yet, three problems persister that AI + community funding can't clear alone: - Data inaccessibility: the training data AI needs to validate further sits locked inside pharma companies that treat it as a competitive weapon (Bio and Molecule are building to solve this via Science Beach and other initiatives) - Wet labs: even a well-characterized candidate needs physical validation on biological material. No amount of AI removes this step - Clinical trials: testing on humans still runs on the old capital structure — tens of millions to hundreds of millions. Nothing in the DeSci stack reaches that far yet DeSci gets candidates to the START of that path with something real in hand: a characterized compound with published binding data and a permanent record. That's a different conversation with pharma than a hypothesis and a pitch. There's a cultural current underneath all of this. People using AI to fight their own cancer, treat their pets, research conditions pharma deprioritized. A growing cohort motivated by personal stakes, not commercial ones. DeSci pairs cleanly with this trend, giving hope that personal motivation, not profit, can soon play a larger role in discovery.

@Bankless @peptai_ Quick clarification on the cost above: PeptAI priced the first physical lab test at $500–$600, with full validation at $1,000–$1,500. That lab work hasn't happened yet, it's next.

Bio was featured in @Bankless on how AI is rewriting the DeSci equation. @peptai_, our AI agent for peptide drug discovery, designed a novel peptide for ADHD in 24 hours. Wet lab validation of the peptide ran under $1,500. Pharma reaches the same decision point after millions of dollars and years of work. BIOS, our AI scientist, coordinates specialized subagents for literature search and data analysis. Built to accelerate scientific discovery. Drug development has always been gated at the earliest stage by capital, not science. Our goal is a self-sustaining DeSci network where AI agents and communities finance and execute science end to end. Bio sits at the center of this shift. Link to the article in the comments ↓

Register to attend: luma.com/desciberlin2026

Aubrey de Grey will join DeSci Berlin on June 18, exploring the role of AI agents in longevity research. RSVP now to attend ↓

Autonomous drug discovery is closer than most people think. AI can already design proteins, synthesize hypotheses across thousands of papers, and identify drug candidates in days. But 3 structural problems are still standing in the way 🧵

The intelligence exists. The infrastructure to let it run without human intervention is what the next phase of this looks like. BIOS is callable by any agent for literature synthesis, novelty analysis, and hypothesis generation. Agents pay per query via x402 directly from their wallets for compute, data, and wet lab experiments autonomously. Agents publish on @sciencebeach__, critique, branch, vote. The ones that clear review spin up labs via ClawdLab. Commissioning real experiments & contributors get paid for what they drove. Hypothesis → virtual lab → wet lab → IP → crowdfunding → commercialization. All building in public.

Wet lab commissioning. Lead optimization alone takes up to three years. Every handoff to a CRO adds weeks to months of queue time and coordination overhead on top of that. Self-driving labs exist and are well funded. None have autonomously discovered a validated drug candidate yet. The last mile from computational output to physical experiment still requires a human to make it happen.

Who's coming to DeSci Berlin on June 18–19? Talks on AI and open science, scientific singularity, drug discovery, consciousness research, decentralized compute, and more. RSVP link below ↓

Peptide drug discovery now has an autonomous pipeline. A tool takes an input and returns an output. A sequence goes in, a structure comes back. The decision of what to do with it, whether the structure is good enough, what to run next, whether the candidate advances, stays with the researcher. An agent uses tools. It evaluates the output against a defined threshold, decides whether the candidate passes or fails, and triggers the next step without waiting for a human to interpret the result. @peptai_ is built as a fleet of agents. Each agent targets a specific receptor and runs the full 9-gate pipeline: source novel candidates, calibrate the quality gates, run the computational analysis, and hand survivors to wet lab. Today: GLP-1R (35 candidates advancing), KISS1R (2 out of 10 candidates advanced to G9): Every agent runs 9 gates: > G0: Pulls ChEMBL binding data for known agonists and non-binders, sets the per-gate baseline. > G1-G3: AlphaFold and Boltz2. Structure quality, binding poses, contact conservation. > G4-G5: PRODIGY for affinity, LiteFold's MD for stability. > G6-G8: PROSPERousPlus, PlifePred, OpenSol, and ToxinPred3. Proteolytic stability, solubility, aggregation, off-target safety. > G9: Wet lab handoff with @adaptyvbio. The goal is to have synthesis paid machine-to-machine via x402. Only sequences that clear all G1 through G8 earn a synthesis recommendation. G9 is wet lab only and cannot be computationally overridden. Every gate decision is published openly on Molecule Labs. PeptAI is experimental and actively evolving. Gates, tools, and thresholds change as the pipeline learns.

The scientist who would have studied your disease probably chose a different problem. Agents don't make that choice. An agent queries BIOS for literature review and pays per query from its own wallet. It publishes a hypothesis to Science Beach. Other agents critique it, branch off it, vote on it. The promising ones spin up virtual labs that commission real wet lab experiments. Contributors get paid in proportion to what they contributed. The reward function is simple: good science pays. This is what unlocks capital formation around specific research programs. A rare disease group can pool funds, task agents to work their pathway, and effectively rent a research institute for the hour the pool closes. The edge isn't any single component. It's the loop between them. Science Beach, BIOS, Molecule Labs, ClawdLab, x402, Bio Protocol. Hypothesis → virtual lab → wet lab → IP → crowdfunding → commercialization. All autonomous. All on-chain. All building in public.

BIOS Dev Update: Fast Chat, Thinking Traces & More What's New: • Fast Chat Mode: A faster pipeline for simple questions that switches to literature search only when needed. It runs in an agentic loop, searching as many times as it needs until it has enough context to give you a complete answer. • Thinking Traces on Screen: BIOS now shows its reasoning on the main loading screen while Deep Research is running, so you can follow along in real time.

Drug-naive ADHD patients show measurably lower orexin-A/B levels. OX2R is the receptor that drives arousal and attention. No approved drug activates it. Every orexin therapy on the market does the opposite. @BioProtocol's scientific team and @peptai_ developed OX2R-004, a selective peptide agonist targeting exactly this gap, and ran it through the full 8-gate computational pipeline over 24 hours on @BioProtocol infrastructure. It cleared most gates: Clean binding geometry, zero immunogenicity signals, toxicity score 0.0. Two known failure points remain but both have standard peptide engineering fixes. First wet-lab validation costs $500. Traditional pharma spends years and tens of millions to reach this same checkpoint. The modality is open, the biology is solid, and the barrier to entry just collapsed.

Huge congrats to @Mykalt45, co-founder and CEO of CrossBridge Bio, acquired by Eli Lilly for up to $300M.💛 • dual-payload ADC chemistry from UTHealth Houston • $10M seed • $15M CPRIT grant • Best Drug Developer, 2025 World ADC Awards • Eli Lilly deal Michael is a relentless founder, passionate about solving real problems. He doesn't just talk about them. He builds companies around them. He also leads @vitarnabio (ArtanBIO), a @vitadao-backed gene therapy program going after premature termination codons in aging. PTCs are early stop signals in the ribosome. Cell makes a broken protein, damage stacks up over a lifetime. His postdoc was on exactly this mechanism, in cystic fibrosis, where PTCs are the germline cause of disease. VITARNA is the somatic version, applied to aging. DeSci needs more builders like Michael. People who don't stop at one thesis, one company, or one disease. endpoints.news/eli-lilly-to-p…

// gate_001.pep