Lukas Milles

Research Group "Biomolecular design"

De novo protein design, deep learning, (single-molecule) biophysics, protein mechanics and folding, high throughput protein characterization
 

Our research group works at the intersection of de novo protein design, deep learning and fundamental biophysics of protein function. We aim to deconstruct biological function by reconstructing such function de novo. Using deep learning based protein design, we construct new-to-nature proteins that incorporate desired functions to better understand their biological mechanism. In particular we are interested to create synthetic autocatalytic enzymes that can covalently target selected epitopes and self-strengthening catch-bonds, both inspired by bacterial pathogen mechanisms. We rely on our recently developed technology for medium/high (96-192 designs/day) throughput (single-molecule) biophysical characterization.  Thus, we can rapidly take proteins from an idea to a blueprint on the computer and finally to the wet lab, and back to study the differences between native and computationally designed protein sequences.