The structural basis and the biological importance of protein communication

Our group consists of interacting scientists that are interested in improving our basic understanding and manipulation of interactions between proteins, using a combination of computational and experimental approaches.

This embraces different levels of resolution and scale: starting from the basic atom - level details of interactions; continuing to prediction and characterization of specific interactions; and finally addressing the ultimate question of their role within the context of a cell and a whole organism.

A main focus of our research is centered around peptide-mediated interactions: protein communication mediated by short motifs. We are interested in understanding the details of individual interactions, and how these are integrated into a complex regulatory signal.

Our computational tools include the structure-based computational prediction and manipulation of specific interactions using the Rosetta modeling framework, analysis of evolutionary signals hidden in sequences, and large-scale integration of this data by machine-learning approaches. We complement these with experimental validation of binding using traditional biophysical assays as well as large-scale display techniques.

Our peptide docking protocol Rosetta FlexPepDock has pioneered the way we model the structure of peptide-protein interactions, and has stimulated the further development of additional approaches. We have also developed approaches for global docking, where only the sequence of the peptide and the receptor structure is known (or only its sequence): These approaches are based on the concept that peptide binding can be seen as complementation of the receptor structure, and therefore, protein structure modeling approaches can be applied to peptide docking. Three approaches are possible: (1) Piper-FlexPepDock: from the peptide side, we can approximate the peptide conformational ensemble by a set of fragments (using the Rosetta Fragment picker), and decouple the search for the peptide conformation from the search of its location on the receptor, (2) PatchMAN: from the receptor side, we can learn how fragments contact the receptor by looking for local matches in solved structures (using MASTER), add the sequence of the peptide onto these fragments, and further refine them using FlexPepDock, and (3) Alphafold: We can dock and fold the peptide-receptor complex as one unit using Alphafold. This is the fastest way to generate a complex. We are currently calibrating a pipeline that makes optimal use of the different protocols.

Our protocols are available as servers, and as part of the Rosetta distribution.

More...

Go to old lab page

We thank our generous funding sources