MassiveFold: unveiling AlphaFold’s hidden potential with optimized and parallelized massive sampling

Abstract Massive sampling in AlphaFold enables access to increased structural diversity; in combination with its powerful confidence ranking, this unlocks elevated modeling capabilities for monomeric structures and foremost for protein assemblies. However, the approach struggles with GPU cost and data storage. MassiveFold removes these restraints as an optimized and customizable version of AlphaFold that runs predictions in parallel, offering the full benefit of enhanced sampling for protein structure and assembly modeling.