Protein language models (pLMs) have demonstrated success at generating functional proteins across vast sequence spaces but lack the ability to design high-fitness variants on demand. Here, we iteratively guide pLMs toward user-defined objectives by applying reinforcement learning (RL). We demonstrate that RL can steer pLMs toward various protein properties, such as topologies or binding affinities, in a few iterations through long evolutionary trajectories. We apply our framework to the design of epidermal growth factor receptor (EGFR) binders, achieving a 26-fold increase in binding affinity in two iterations.
@article{arxiv.2412.12979,
title = {Guiding Generative Protein Language Models with Reinforcement Learning},
author = {Filippo Stocco and Maria Artigues-Lleixa and Andrea Hunklinger and Talal Widatalla and Marc Guell and Noelia Ferruz},
journal= {arXiv preprint arXiv:2412.12979},
year = {2025}
}
Comments
28 pages including main text and supporting information