Nature's supply is fragile
Valuable molecules are extracted from plants, where yields are almost always low. Supply depends on harvests and long chains that geopolitical or supply-chain instability can threaten.
Computational biosynthesis
Routes to molecules the
world can't make yet.
Give us a target molecule. We compute a route to build it in a microbe, engineer the enzymes its hardest steps need, and score every candidate. Then we take the best ones to the bench and hand you back a strain that makes it.
- 1.5M+
- compounds, one namespace
- 77,107
- reactions, one searchable map
- Galantamine
- lead program
The problem
Making the right molecule is slow, costly, and mostly guesswork.
Chemistry hits a wall
Complex molecules take dozens of steps, harsh reagents, and low yield. For many targets, no economical route exists.
Bio-routes are found by trial
Engineering a microbe means years of guessing pathways and enzymes, then testing thousands of variants by hand.
The platform
From target molecule to testable shortlist, in software.
The expensive part of discovery moves out of the lab and into computation. The questions an expert team would ask, across the whole search space, in a fraction of the time.
-
01
Find the route
Search biosynthetic routes to the target through a microbial host, working backward from product to feedstock.
-
02
Locate the hard steps
Flag the reactions with no known efficient enzyme. These bottlenecks decide whether a route is real or just a diagram.
-
03
Design the enzymes
Generate enzyme variants built to catalyze those exact steps, tuned to the reaction they have to run.
-
04
Score and rank
Predict which variants are likely to catalyze, and return a ranked, scored shortlist.
The output is a short, ranked list of what to build. We build the top candidates and deliver the strain that works.
Lead program
Galantamine
Galantamine treats the symptoms of Alzheimer's. Today it comes from daffodil extraction or long chemical synthesis. Both are constrained and expensive.
It's the ideal first proof point: high value, a genuinely hard synthesis, real demand. We're designing the route and the enzymes its hardest step needs.
- Established clinical use and demand
- Supply constrained by extraction and synthesis
- A synthesis hard enough to prove the platform's edge
Our approach
Software narrows thousands of options down to a handful worth building. Then we go to the bench and find out which ones actually work.
Right about what to test
Designing new catalytic activity is hard, and we don't pretend otherwise. The edge is being right about which few candidates deserve a wet-lab slot.
Proven at the bench
Accuracy is earned in the design-build-test loop and shown by recovering known results blind. We measure it against experiment before we claim it.
Why now
For the first time, biology is something you can design.
Models predict function
Protein-structure and reaction models can now estimate whether a designed enzyme works, before it's made.
Compute caught up
Searching routes and scoring thousands of variants now runs on infrastructure a small team can afford.
One integrated system
The edge is the whole pipeline: route search, enzyme design, and scoring wired together so a target goes in and a ranked answer comes out.
Get in touch
The design layer for biomanufacturing.
For investors and partners ready to look closely: the deck lays out the platform, the galantamine program, and the roadmap from computed route to producing strain.
Or email us at hello@novanoro.com