AI foundation models: funding of the PROFOUND project
Proteins are fundamental molecules of life that control all processes inside living systems. Emphasized by the 2024's Nobel Prize in Chemistry, the ability to design proteins has unlocked tremendous potential for many applied areas in synthetic chemistry and medicine. Today, AI can be used to predict the three-dimensional structure of the protein from their gene sequence. Like nanomachines, proteins are highly dynamical and their structural flexibility is often crucial for their biological function. However, current state-of-the art AI models such as AlphaFold are limited to the prediction of static protein structures. PROFOUND aims at revolutionizing protein design by tackling this limitation with the creation of a new AI model that can predict dynamical protein models from large-scale molecular dynamics data. This approach will allow the design of programmable molecular machines that will lead to innovations in smart therapeutics, sustainable materials, and next-generation biotechnologies.
This project is funded for three years and will involve the following participating research centers: Forschungszentrum Jülich (IBG-4 and JSC), Helmholtz Munich, Helmholtz-Zentrum Berlin, and Helmholtz-Zentrum Dresden-Rossendorf.