SaprotHub: Making Protein Modeling Accessible to All Biologists

Abstract Training and deploying deep learning models pose challenges for users without machine learning (ML) expertise. SaprotHub offers a user-friendly platform that democratizes the process of training, utilizing, storing, and sharing protein ML models, fostering collaboration within the biology community—all without requiring extensive machine learning expertise. At its core, Saprot is an advanced, foundational protein language model. Through its ColabSaprot framework, it supports potentially hundreds of protein training and prediction applications, enabling the co-construction and co-sharing of these trained models. This enhances user engagement and drives community-wide innovation.