Work Packages

WP1: Enzyme engineering and understanding via state-of-the-art wet- and dry-lab approaches


Objective 1: To combine both classical and AI/ML-guided directed evolution to engineer multiple enzyme properties, focusing on data generation using high-throughout screening assays such as automation, genetic selection systems, and microfluidics.

Objective 2: To investigate conformational dynamics using MD simulations to guide engineering efforts. Hybrid approaches combining data-driven and mechanistic knowledge can lead to important insights into enzyme function.

Objective 3: To develop oxygenases with improved activity, stability, and/or selectivity that will be tested in small & bench-top bioreactors together with WP2, whereas the data generated will serve to refine AI/ML tools developed by WP3.

WP2: Improve bioprocess design and control by combining cutting-edge experimental and computational tools


Objective 1: To understand how the instability issues associated with oxygenases limit bioprocess development but can guide an optimal provision of O2 or H2O2 to reach high volumetric productivities. 

Objective 2: To investigate the multi-scalability potential of oxygenases with varying levels of complexity by combining experimental and computational approaches and obtain products in g-scale.

Objective 3: To demonstrate how AI/ML offers the possibility to design optimised biocatalytic processes in a streamlined, data-driven approach based on global, state-of-the-art knowledge, using real-time analytics for generating the maximum amount of process data for bioprocess design.

WP3: Develop novel AI/ML tools to support the integrated design of enzymes and bioprocesses in biocatalysis.


Objective 1: To optimize enzyme function (stability, activity, and/or selectivity) by using existing databases and data from WP1 and provide expertise for coupling state-of-the-art AI/ML methods with directed evolution experiments.

Objective 2: To address issues of slow throughput and high computational costs by developing AI/ML tools for streamlining Molecular Dynamics and Empirical Valence Bond simulations and their systematic analysis using existing data and data from WP1.

Objective 3: To enable the development of down-scale models to increase the predictability of the performance of engineered enzymes under industrially relevant process conditions by developing AI/ML tools for an integrated bioprocess design using process and kinetic data from WP1-2.