A biomanufacturing company nearly tripled P450 enzyme activity in three Cradle rounds after plateauing.
After 10 rounds and 1,200 P450 variants, rational design had hit diminishing returns. Cradle's platform was trained on the company's historical data and proposed mutations combined with human expertise over three rounds of 96-192 sequences. The page reports activity nearly tripled in 3 rounds, a 4x faster rate of improvement, and activity more than doubled in the first two rounds.
Cradle is an AI protein engineering software company whose platform designs and optimizes proteins, antibodies, enzymes and peptides for biopharma and industry.
This sits in the Design & Stress-Test stage of the AI use-case lifecycle: Named examples from the tools pharma teams use to scope, prototype and benchmark an AI use case before committing to build it — data-catalog, sandbox and evaluation platforms.
Published by Cradle on its own site — read the original case study on cradle.bio ↗
Note: Industrial enzyme use case rather than pharma.
Related use cases
- 01The Keasling Lab used Cradle to re-engineer plant enzymes for fermentation-based QS-21 vaccine adjuvant production.
- 02A top-20 pharma optimized three stalled late-stage peptide programs with Cradle in one round each.
- 03A top-20 pharma improved the thermostability of a S. aureus vaccine antigen in a single Cradle round.
- 04A top-50 pharma delivered 10 IgG candidates meeting potency and developability criteria after earlier campaigns failed.
- 05Novo Nordisk Transforms Clinical Trials With AI
- 06How Regeneron is discovering new treatments with AI