Computational Protein Engineering

What we try to achieve
We aim to understand how proteins work at the molecular level, while engineering them to perform specific tasks. However, engineering a protein is a complex task leading to low success rates caused by tradeoff effects, diminishing returns and non-linear epistatic interactions. We develop novel computational methods to optimize protein and enzyme function(s) efficiently, while validating such methods in the wet lab.
 
Why our research is important and how it can be used
Proteins are one of the most versatile building blocks of life involved in the transport of nutrients, transfer of energy, stability of molecular structures and conversion of matter. For example, enzymes can convert waste such as CO2 into value-added compounds like natural products, biochemicals and biomaterials.
 
How we achieve our aims – methods, tools, technologies
We use both computational and experimental methods in protein engineering including directed evolution, one of the most successful technologies awarded a Nobel Prize in 2018 for its power to evolve unrestricted protein functionality. From the computational side, we use machine learning, bioinformatics, and molecular dynamics simulations. Experimental tools include not only high-throughput systems that can generate big data like genetic selection in microbes and microfluidics based on water-in-oil emulsions, but also biochemistry, enzymology, and biophysics. Wet-lab efforts are complemented by state-of-the-art methods in analytical chemistry, automation, and robotics, often in collaboration with the Biofoundry team at DTU Biosustain.