CIS – “Get to know your neighbors” Seminar Series
“Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning“
Prof. Bruno Correia, Tenure Track Assistant Professor , Laboratory of Protein Design & Immunoengineering (LPDI)
Monday 1st February 2021 – 3:15 – 4h15pm (CET)
Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning
Predicting interactions between proteins and other biomolecules solely based on structure remains a central challenge in biology. A high-level representation of protein structure, the molecular surface, displays patterns of chemical and geometric features that fingerprint a protein’s modes of interactions with other biomolecules.
The underlying hypothesizes of our work is that proteins interacting with similar molecules may share common fingerprints, independent of their sequence and overall structural fold. These structural fingerprints are difficult to grasp by visual analysis but may be learned from large-scale datasets.
I will discuss MaSIF (Molecular Surface Interaction Fingerprinting), a conceptual framework based on a geometric deep learning method, to capture structural fingerprints that are important for specific biomolecular interactions. I will also present current developments of the framework to make it end-to-end differentiable. Finally, I will describe our approach based on MaSIF to design de novo protein-protein interactions, which may have important applications in the development of new protein-based drugs.
We anticipate that this conceptual framework will lead to improvements in our understanding of protein function and design.
Throughout his PhD and postdoctoral studies he was trained in world-renowned laboratories and institutions in the United States of America (University of Washington and The Scripps Research Institute). Very early in his scientific career he found out his fascination about protein structure and function. His PhD studies evolved in the direction of immunogen design and vaccine engineering which sparked his interest in the many needs and opportunities in vaccinology and translational research. His efforts resulted in an enlightening piece of work where for the first time, computationally designed immunogens elicited potent neutralizing antibodies. During his postdoctoral studies he joined a chemical biology laboratory at the Scripps Research Institute. In this stage he developed novel chemoproteomics methods for the identification of protein-small molecule interaction sites in complex proteomes.
In March 2015, he joined the École Polytechnique Fédérale de Lausanne (EPFL) – Switzerland as a tenure track assistant professor. The focus of his research group is to develop computational tools for protein design with particular emphasis in applying these strategies to immunoengineering (e.g. vaccine and cancer immunotherapy). The activities in his laboratory (Laboratory of Protein Design & Immunoengineering) focus on computational design methods development and experimental characterization of the designed proteins. His laboratory has been awarded with 2 prestigious research grants from the European Research Council. Lastly, he has been awarded the prize for best teacher of Life sciences in 2019.