Bringing a new drug to the European market takes at least 10 years and 2.5 BEUR of R&D effort. Computational methods significantly shorten this journey but they require knowledge of the structure and interactions of the involved biomolecules - most often proteins. In recent years, a tremendous progress has been made in the field of a single protein 3D structure prediction. However, predicting protein assemblies -the most crucial step - still remains very challenging. The aim of this IF project is to revolutionise protein complexes prediction methods. This will be achieved first by developing novel, effective and fast approaches for the calculation of the vibrational entropy, key to protein-protein docking mechanisms. Then, in an innovative and multi-disciplinary approach, the Experienced Researcher (ER) aims to combine advanced physics-based models with machine learning methods using data from structural and sequence databases. Finally, this project will link all the pieces together and release them in the form of a web-server in order to allow the community to benefit from the results of this research.
The ER will carry out the fellowship in the Centre National de la Recherche Scientifique - CNRS in Grenoble, France. CNRS carries out research in all scientific fields of knowledge and the Supervisor is a renowned expert in data science, computing, and software engineering. Through a well-thought two-way knowledge transfer and training plan, this project will benefit both the host institution and the ER in terms of scientific knowledge, network and open the path for new applications to potentially exploit at the European or global level. The project will also place the ER as a highly visible researcher in the field and ideally set her as a valuable resource for European industrial actors.