Google DeepMind's AlphaProtein Novo designs new-to-nature enzymes from scratch (piperidine synthesis, DEHP degradation), with Frances Arnold's lab
On Oct 5, 2026 Google DeepMind posted two bioRxiv preprints on AlphaProtein Novo, a diffusion system that co-generates enzyme structure and sequence around a target reaction. The authors say it is the first time de novo enzyme design outperforms mining natural sequences on hard reactions: it produced nitrene transferases that make chiral piperidines (a common drug motif) and DEHP-degrading enzymes that work under denaturing conditions. Code is on GitHub (Apache 2.0); weights are under separate terms.
Key facts
- Preprint 1: 'Designing enzymes for new-to-nature chemistry and non-natural substrates with AlphaProtein Novo' (Wu, Abramson, Frerix et al.; Google DeepMind, Caltech, Pitt), bioRxiv 10.64898/2026.10.01.756017
- Claim: 'the first time de novo enzyme design can outperform natural sequence mining' on challenging reactions; competitive activity on benchmark reactions; key ingredient: mechanism-inspired metrics based on AlphaFold 3 predictions
- Preprint 2 (with Frances Arnold's Caltech lab and Peng Liu's Pitt group): 'Design-Evolution Synergy Enables New-to-Nature Enzyme Chemistry', de novo hemoproteins for regio- and enantioselective C–H amination to chiral piperidines, combining motif scaffolding, redesign and directed evolution (bioRxiv 10.64898/2026.10.01.754751)
- DEHP (a phthalate plasticizer pollutant) degraders that remain active under denaturing conditions that inactivate natural enzymes
- Repository: google-deepmind/alphaprotein-novo, a diffusion pipeline with LigandMPNN redesign and AlphaFold 3 scoring; benchmarks include Kemp eliminase, serine esterase, DEHPase, carbene and nitrene transferases; code Apache 2.0, weights under separate terms of use
- Announced by Pushmeet Kohli on X (~31k views, Oct 5), crediting Jue Wang, Zachary Wu, Josh Abramson and the Arnold and Liu groups
- Secondary claim (Startup Fortune): one enzyme makes piperidine '99 times faster than current industrial methods'; not found in the abstracts, unverified
Science result
- Field
- biology / protein design / enzyme engineering
- Problem
- De novo design of enzymes for custom (including new-to-nature) reactions
- Result
- Generative diffusion system that designs functional enzymes from scratch, beating natural-sequence mining on hard reactions; nitrene transferases for chiral piperidines and heat/denaturant-tolerant DEHP degraders, lab-tested
- AI system
- AlphaProtein Novo, AlphaFold 3
- Human role
- Human-led with AI tools: DeepMind designs, wet-lab testing and directed evolution by Caltech (Arnold lab) and collaborators
- Verification
- Preprints (bioRxiv), lab-validated by the authors; not peer-reviewed
- Status
- pending
- Why surprising
- De novo enzyme design, long a 'holy grail', is claimed to beat nature's own sequences as a starting point for hard chemistry.
What happened
DeepMind extended its protein work (AlphaFold for structure, AlphaProteo for binders) to enzymes. AlphaProtein Novo generates a protein's backbone and sequence together around a catalytic motif, redesigns sequences with LigandMPNN and filters designs with AlphaFold 3-based, mechanism-aware metrics. With Frances Arnold's lab, designs were further improved by directed evolution into efficient, stable enzymes for chiral piperidine synthesis.
Why it matters
Most industrial enzymes start from natural proteins and are improved by mutation. A generator that beats natural starting points for non-natural chemistry would open enzyme routes to drug intermediates and pollutant clean-up that nature never evolved. Results are preprints; claims such as "99x faster than industrial methods" come from secondary coverage and are not verified here.
Changelog
- 2026-10-06: created
People
Related posts (1)
- Pushmeet Kohli original ↗ Pushmeet Kohli @pushmeet · x · 2026-10-05
Cited as a source by: 2026-10-05-deepmind-alphaprotein-novo-enzyme-design
Related events
- Claude agents discover a novel CRISPR-like enzyme system; Anthropic reveals its own biology wet lab ★★★★
- Google DeepMind introduces SynthID Bio, watermarking for AI-designed proteins and DNA (Nature paper, open code and weights) ★★★
Sources (5)
- paperbioRxiv: Designing enzymes for new-to-nature chemistry and non-natural substrates with AlphaProtein Novo
- paperbioRxiv: Design-Evolution Synergy Enables New-to-Nature Enzyme Chemistry
- codeGitHub: google-deepmind/alphaprotein-novo
- officialPushmeet Kohli on X: AlphaProtein Novo preprint
- pressStartup Fortune: DeepMind unveils AI system that designs enzymes no living cell ever made
id: 2026-10-05-deepmind-alphaprotein-novo-enzyme-design · updated 2026-10-06 · open in the interactive timeline