Carles Navarro

Learning to make Neural nets learn chemistry. 🧠🤖

    • Bio

    In the cosmic realms of computation, Carles Navarro was renowned as the maestro of deep neural networks. Trained in the arcane arts of chemistry, he ventured into the labyrinth of code, aspiring to become a software architect. With every epoch in his training cycle, he drew inspiration from the alchemical fusion of science and technology. As he delved deeper into the mysteries of physics and the intricacies of software engineering, the world watched in awe. Journey with him as he bridges the chasm between molecules and matrices, in a universe where science, software, and endless curiosity intertwine.

    History

    2022 -
    Industrial PhD: Industrial PhD at Acellera Labs. Using Generative AI for drug discovery.
    2020 - 2022
    Msc Bioinformatics for the Health Sciences: Worked at Computational Science Laboratory with Gianni de Fabritiis. Differentiable Molecular Simulations for protein folding. Developed a new methodology for training a Graph Neural Network with simulations to learn the coarse-grained force field parameters.
    2019 -
    Bsc Physics: Studying physics for fun.
    2015 - 2019
    Bsc Chemistry: Discovered that I liked science but not wet labs. So now I try to do science with computers.

    Publications

    ACEGEN: Reinforcement learning of generative chemical agents for drug discovery
    J. Chem. Information and Modelling 2024
    Albert Bou, Morgan Thomas, Sebastian Dittert, Carles Navarro, Maciej Majewski, Ye Wang, Shivam Patel, Gary Tresadern, Mazen Ahmad, Vincent Moens, Woody Sherman, Simone Sciabola, Gianni De Fabritiis
    On Machine Learning Approaches for Protein-Ligand Binding Affinity Prediction
    arXiv:2407.19073 2024
    Nikolai Schapin, Carles Navarro, Albert Bou, Gianni de Fabritiis
    Top-down machine learning of coarse-grained protein force-fields
    J. Chem. Theory Comput. 2023
    Carles Navarro, Maciej Majewski, Gianni de Fabritiis