Custom service

Computer-aided Enzyme Design

Designing an active site around the chemistry you want.

Chemistries: Hydrolysis, transfer, redox and bond-forming reactions; cofactor-dependent chemistry supportedTools: Rosetta design suite, docking, molecular dynamics, QM reference calculationsScreening scale: 10–100 designed variants per cycle

De novo enzyme design is one of the hardest problems in the field, and the honest framing is that activity comes from catalysis, not from binding. A designed pocket that positions the substrate well may still be a poor enzyme because it does not stabilise the transition state. The design process therefore starts from the transition state, not from the substrate.

The workflow is iterative by nature. The desired chemistry is specified — bond formed or broken, cofactor used, stereochemistry required — and a theozyme, an ideal active-site geometry that stabilises the transition state, is built first. Rosetta or comparable design tools then fit protein scaffolds around that geometry, and each designed sequence is scored for pocket packing, folding stability and solubility before anything is made.

Most designed enzymes are weak on the first pass, and the numbers say so plainly: kcat/KM improvements of orders of magnitude come from the experimental optimisation cycles that follow, not from the initial design. Directed evolution around a designed active site is part of the workflow, not an admission of failure — the design provides the starting point, and evolution provides the catalytic proficiency.

Computer-aided design fits a defined reaction that no natural enzyme performs, or an existing enzyme whose specificity must be redirected at the active site.

It is the wrong tool for improving an enzyme's thermal stability — that is stability engineering — and for cases where a natural enzyme with the desired activity probably exists and simply has not been found. Sequence and metagenome searching is the cheaper first move.

Every design is delivered with its kinetic parameters, the verification that the measured activity is the intended chemistry, and the design rationale connecting the active-site geometry to the outcome. Designs that showed no activity are documented, since they inform the next cycle.

Service Highlights

  • Transition-state first. Theozyme design precedes scaffold selection, because stabilising the transition state is the actual design goal.
  • Scaffold ranking. Candidate scaffolds scored for pocket packing, predicted stability and solubility before synthesis.
  • Cofactor-aware design. PLP, metal ions, flavins and other cofactors modelled explicitly, including their pre-organised positioning.
  • Kinetic acceptance criteria. Designs judged on kcat, KM and kcat/KM against the specified target, not on substrate binding alone.
  • Evolution handoff. Weak designs enter a directed evolution cycle focused on the active-site second shell.
  • Mechanistic verification. Activity confirmed as the intended chemistry by substrate and product analysis, not by a generic readout.

How the work runs

  1. Reaction specification

    Bond chemistry, cofactor, stereochemistry and target kinetics defined; feasibility assessed against known precedents.

  2. Theozyme design

    Ideal active-site geometry built around the transition state, with catalytic residues positioned.

  3. Scaffold search and design

    Protein backbones fitted to the geometry; sequences designed and scored.

  4. Expression and screening

    Designed variants expressed and screened for the intended activity at single-point and combination level.

  5. Kinetic characterisation

    Active variants characterised by kcat, KM and mechanism verification.

  6. Optimisation cycle

    Directed evolution on second-shell residues; iterated until the kinetic target is reached.

Specifications & timeline

Chemistries
Hydrolysis, transfer, redox and bond-forming reactions; cofactor-dependent chemistry supported
Tools
Rosetta design suite, docking, molecular dynamics, QM reference calculations
Screening scale
10–100 designed variants per cycle
Kinetics
kcat, KM, kcat/KM by steady-state assays
Timeline
4–10 weeks per design and optimisation cycle

Related Products

Next step

Tell us the target, we will tell you what is feasible

Send the sequence, assay format and sensitivity you need. You will get a technical assessment — recommended platform, timeline and the risks worth knowing before you commit budget.