De Novo Enzyme Designing

Computational de novo enzyme design

De novo enzyme design aims to computationally create new enzymes, or substantially re-engineer existing ones, to catalyse target reactions that may not exist in nature. By combining catalytic-site (theozyme) design, scaffold selection, deep-learning protein design and molecular modelling, it is possible to build biocatalysts with novel or improved activity, specificity and stability.

We support enzyme engineering and design projects from concept through to computationally validated candidates, ready for experimental testing.

Our Services

• Active-site (theozyme) and catalytic-mechanism design
• Scaffold selection and de novo backbone generation
• Deep-learning sequence design for chosen scaffolds
• Mutation prioritisation for activity and thermostability
• Substrate docking and binding-mode analysis
• Molecular-dynamics validation of designs
• Stability and solubility optimisation
• Candidate ranking for experimental screening

These approaches power applications in green chemistry, industrial biocatalysis, biosynthesis and therapeutic enzyme development.

What you receive

• Designed enzyme sequences with structural models and predicted catalytic geometry
• Substrate docking and binding-mode analysis for each candidate
• Molecular-dynamics validation of active-site stability over the simulated timescale
• Ranked mutation lists for activity, selectivity and thermostability
• Expression and solubility assessment for the intended host
• A written report and the complete computational record

Two routes to a better enzyme

Most projects are one of two kinds, and they differ substantially in risk.

Engineering an existing enzyme — improving activity, thermostability, solvent tolerance or substrate scope on a known scaffold. Success rates are good and the computational work is well validated: mutation scanning, stability prediction, docking of the intended substrate and simulation of the resulting variants.

Designing catalysis into a new scaffold — building a theozyme describing the ideal catalytic geometry, finding or generating a scaffold that can present it, and designing the sequence around it. This is far more ambitious. It has produced real de novo enzymes, but usually with activity well below natural counterparts, and directed evolution is normally needed afterwards to reach useful turnover.

We will tell you which category your project falls into and what that realistically implies for timeline and success probability, before work begins.

How the design work runs

• Reaction and mechanism analysis, including transition-state modelling where the chemistry requires it
• Catalytic-site design or identification of positions controlling the existing mechanism
• Scaffold selection from structural databases, or de novo backbone generation
• Deep-learning sequence design, with the catalytic residues constrained
• Structure prediction and validation of the designed sequence
• Substrate and transition-state-analogue docking to confirm productive binding geometry
• Molecular dynamics to test whether the active-site arrangement persists rather than collapsing
• Stability, solubility and expression-liability filtering, then ranking for experimental screening

Applications

• Industrial biocatalysis and green chemistry replacements for harsh chemical steps
• Biosynthetic pathway engineering and precursor production
• Plastic and pollutant degradation
• Therapeutic and diagnostic enzymes
• Enzymes tolerant of non-natural solvents, temperatures or pH