De novo longevity chemistry,measured — not promised.
Target × tissue × hallmark. Every candidate ships a route you can run and an interval on every score.
2D structureCOC1=C/C(=C\C(=O)C(=O)Nc2nc(O)c3cc(C)ccc3n2)OC(O)=C1Assembled from real chemistry — so the route exists before the molecule does.
Most generators emit a structure and guess the synthesis afterwards. We build from the reaction, so the route is not a prediction.
The proof, in three numbers.
Backed by a methods preprint — including the number where we fail.
Recovers known binders from decoys on a scaffold-disjoint held-out split.
Level with the strongest unconstrained generators (0.94–0.95) — and ours stay makeable.
The lifespan prior is weak but real (ρ 0.182) — always shipped with its interval.
Three steps. Seconds on CPU.
Pick a target
Eleven, ranked by human genetic evidence.
→Set the tissue
One tissue — or whole-body, pan-aging.
→Get candidates
Verified route and an interval on every score.
Built for teams hunting real longevity chemistry.
Longevity biotechs
Makeable candidates before you spend on synthesis.
Geroscience labs
A validated target → testable molecules.
Drug-hunting teams
Clean structures, ADMET, a ranked shortlist.
Designed on three axes — target, tissue, hallmark.
We fix the biology first — not a molecule in a vacuum.
Target
Eleven validated longevity mechanisms — CD38, mTORC1, IGF1R, Bcl-2, NAMPT and more.
Tissue
Ten tissues — or whole-body, pan-aging.
Hallmark
Twelve hallmarks scored (López-Otín 2023).
What this candidate engages.
The same per-hallmark score every molecule carries inside the platform.
Real molecules. Real scores. Real structures.
Top candidates from a real whole-body mTORC1 run — exactly as produced.
Built for aging biology, not retrofitted from oncology.
Built for aging
Twelve hallmarks. Not an oncology retrofit.
Honest uncertainty
Every score ships its interval and its accuracy.
CPU, seconds
No GPU. Every candidate carries a real route.
11 targets · 10 tissues
One pathway, or whole-body.
The pathways longevity actually runs on.
Design against one mechanism, or switch to whole-body mode and optimise across all ten tissues at once.
NAD⁺ → SIRT1 / PGC-1α metabolic axis
Chromatin guardian; NF-κB co-repression
Rate-limiting enzyme of NAD⁺ salvage
Senolytic clearance of senescent cells
Autophagy gate; canonical longevity axis
Cellular energy sensor; ADaM-site activation
Suppress the SASP / inflammaging
Silence the senescence secretory phenotype
RAS–MEK–ERK inflammaging axis
Master antioxidant-response regulator
IGF-1 / insulin (daf-2) longevity axis
We don't ship a score we can't measure. Where the signal is strong, we say so. Where it's weak, we show the interval — and let your chemists decide.
A field burned by overpromising — we measure, or we don't ship.
Bring your target. We'll bring the chemistry.
Novel, makeable candidates for your longevity program.