Sculpting cavities in alpha-helical bundles to accommodate ligands of different size and shapes

The advancement of AI methods, including RoseTTAFold and AlphaFold (structure prediction from sequence), RFdiffusion (backbone design), ProteinMPNN (sequence design), and more comprehensive tools like Chroma (backbone and sequence design), has made protein structure and function design highly accessible (e.g. through platforms like Colab notebooks) and remarkably successful.However, rational design still offers two significant advantages: it minimizes the need for extensive sequence sampling and enhances understanding.In the Woolfson lab, we employ a hybrid approach that combines rational and computational methods to design functional and complex coiled-coil proteins, beginning with a deep understanding of the sequence design rules for coiled-coil peptide assemblies.