Prof. Lukas Milles

Max-Planck-Institute of Biochemistry & Gene Center Munich

De novo protein design has experienced a machine learning-fueled revolution, vastly expanding our ability to design complex and functional proteins beyond those found in nature. I will discuss recent work at the intersection of de novo protein design, deep learning and fundamental biophysics of protein function regarding both computational design and wet lab biophysical characterization. The work will touch upon control over three levels of protein structure: oligomeric state and symmetry, secondary structure and scaffolding of structural motifs, and finally control over amino acid side chain orientation for catalytic activity.