Millions suffer from diseases we understand but cannot treat.
We open 3-D chemical space with patented synthesis, then AI selects what partners can actually develop.
Why We Exist
The bottleneck in drug discovery isn't biology.
It isn't money. It's the molecules.
For decades, medicinal chemists have searched for drugs in the same crowded library: flat, aromatic, sp²-dominated compounds that are easy to make and easy to buy. The tragedy is not that these molecules are bad, it is that they are nearly all the same. And nature's most important binding sites don't recognise sameness.
The numbers speak to a systemic failure. Not of ambition, not of clinical expertise, but of raw molecular diversity. The drugs we cannot yet make are the drugs we cannot yet give.
"The next breakthrough in medicine will not come from better algorithms or larger datasets alone. It will come from chemistry that has never been done before, from molecular architectures that open doors to binding sites, geometries, and biological interactions that conventional molecules simply cannot reach."
The founding conviction of DAOdiscovery
The Two Problems
Two discovery failures with no chemical answer, until now.
Beyond the patient mission, pharma teams face two structural threats when libraries stagnate or a lead is dropped. Today's flat, well-mapped chemistry offers no defence against either.
Replacement risk
Even when your drug targets one of the 22% of diseases that already have a treatment, a novel scaffold can let a competitor bind the same target more efficiently and capture your market share. Lipitor displaced Mevacor a decade after launch.
No flat, well-mapped library can defend a position once a better-shaped molecule appears.
Deselection cost
Chasing one of the 78% of diseases with no treatment, most discovery programmes still fail. When a lead is dropped, teams need a versatile scaffold that supplies dozens of analogues fast, but such scaffolds are scarce. Projects stall and sunk costs compound.
Structurally distinct rescue scaffolds barely exist, so years of work are simply written off.
The belief
We believe in a healthier world through science and innovation.
We are convinced that diseases without a cure can be cured, and that the cures we already have are not yet realising their full potential. That conviction is why DAOdiscovery exists.
The approach
By opening the under-explored 3-D chemical space with versatile building blocks, guided by AI and digital workflows.
AI explores billions of possible forms and surfaces the fraction worth synthesising, and when one candidate fails, the same versatile scaffold supplies dozens of analogues to keep the programme moving.
The product
Our Diazo Ylide platform: compound libraries, spiropentane scaffolds, and software we keep improving.
We deliver these through compound sales, library licensing, and royalty-bearing co-development, built on peer-reviewed science, a patent family, and years of synthesis expertise.
Platform Scale
Numbers that define our moat, not just the industry's problems.
Macro statistics explain why change is urgent. These numbers explain why DAOdiscovery can deliver it.
How We Think
Four convictions that drive everything we do.
These are not corporate values printed on a wall. They are the beliefs that every experiment, every synthesis, every model run is built upon.
Chemistry is the foundation.
You cannot compute your way to a molecule that doesn't exist. Before AI can navigate chemical space, someone must first open new territory within it. We start with the chemistry, because synthesis is the only proof that matters.
Three dimensions outperform two.
Life happens in three dimensions. Proteins fold, bind, and signal through geometries that flat molecules cannot reach. The future of drug design lives in 3-D chemical space, and today, that space is almost entirely unmapped.
AI is the compass, not the driver.
Generative models can imagine billions of molecules. That's table stakes. What separates a drug candidate from a hallucination is physical reality, whether it can actually be made, tested, and validated. We provide the reality.
The science owes patients an answer.
Behind every untreated disease is a person who was told there is nothing we can do. We believe that answer is unacceptable as long as there remains unexplored chemistry that could change it. That is why we work.
What We Build
From a novel scaffold to a pharma-ready platform.
The vision is realised through three interlocking capabilities, each peer-reviewed, patent-protected, and ready for partnership today.
Novel Chemical Matter
Patented poly-decorated spiropentane scaffolds synthesised via Diazo Ylide, the first general route to a molecular architecture that has never existed before.
AI Pre-selection Engine
From billions of enumerated forms to the highest-priority candidates, filtered by ADMET-AI, QED, and Lipinski Ro5 against your specific disease target.
Pharma-Ready Partnership
Built to de-risk partner pipelines: compound libraries for licensing, co-development, or exclusive access, whether you need to rescue a failing programme or stay ahead of replacement risk.
Spiropentane is where we start, not where we stop. The same Diazo Ylide logic extends to epoxides and additional scaffold families, guided by AI and validated by synthesis at every step. One chemistry platform, many unexplored architectures.
Platform roadmap
Where We Sit
Chemistry-first in a field of AI platforms and catalog vendors.
We do not compete on who imagines the most molecules. We compete on who can make architectures that nobody else sells, and supply the partners who need them.
Catalog & building-block vendors
Insilico Medicine
Generative-AI drug design with its own target discovery and synthesis planning (Chemistry42, Pharma.AI).
It optimises within today's synthesisable, catalog-able space. We extend that boundary, feeding the models scaffolds they cannot otherwise source.
Schrödinger
Physics-based molecular simulation and free-energy prediction, the computational backbone behind many design pipelines.
Their physics scores molecules that can be built. We widen what can be built, with rigid 3-D scaffolds outside the usual chemical space.
Recursion
Biology-at-scale TechBio: high-throughput phenomics mapping cellular responses across vast compound sets.
Its loop still optimises known chemistry. We supply the novel 3-D scaffold classes it cannot reach, a partner rather than a rival.
Exscientia
AI-driven design with automated, closed-loop small-molecule synthesis and precision-medicine focus.
It iterates fast over accessible chemistry. We hand it building blocks no catalogue can supply.
Isomorphic Labs
DeepMind spin-out applying AlphaFold-class models to structure-based drug design at scale.
Structure prediction needs real, makeable ligands to act on. Our chemistry expands the makeable set into unexplored 3-D space.
Iambic Therapeutics
AI-plus-physics discovery engine coupling generative models with automated synthesis and assays.
Like the others, it optimises within today's chemistry. We open new chemistry for it to optimise.
Our Manifesto
We envision a world in which the gap between a known disease target and an effective drug does not take two and a half billion dollars and fifteen years to close. A world in which the chemist's answer to "can we make that?" is not limited by the flat, crowded catalogue of compounds that already exist.
We founded DAOdiscovery because we believe that gap is, at its root, a chemistry problem.Not a computing problem. Not a data problem. A chemistry problem that requires new molecules, molecules with geometries, binding profiles, and spatial properties that no medicinal chemist has ever held in their hands.
The spiropentane scaffold is the beginning of that answer, not the end. Two cyclopropane rings sharing a single carbon. A geometry that nature encoded in three dimensions long before any drug company thought to look for it. Unique bond distances of 270–280 pm. A molecular architecture that fits binding pockets that every flat compound in every library on earth walks past without noticing. From here, the platform extends to epoxides and further scaffold families on the same patented chemistry.
We make those molecules real. AI imagines them. Diazo Ylide builds them. And then we test them, because a drug that only exists in a model has never saved a life.
That is our why. That is why we work.
De-risk your pipeline.
Expand what chemistry can do.
We are seeking pharma partners to pilot our first libraries and investors to scale AI-guided synthesis. Whether you need to rescue a programme, stay best-in-class, or back novel chemical matter, let's talk.