Protein folding stability estimation with an explicit consideration of unfolded states

Thermal folding stability is a critical requirement for the vast majority of proteins. Computational methods suggested to date for the absolute folding stability (Δ G ) prediction – including those driven from protein structure prediction AIs – show clear limitations on reproducing quantitative experimental values. Here we present IEFFEUM, a deep neural network that jointly estimates Δ G and the equilibrium ensemble of folded and unfolded states represented by their residue-pair distance probability distributions. This joint learning considerably enhances the Δ G prediction accuracy against the scenario where Δ G prediction was learned alone. To improve the model, we further extend the dataset compared to previous related works, which includes the Mega-scale small proteins and disordered proteins for training, as well as wild-type natural proteins with sizes up to 900 residues for thorough validation. We show that IEFFEUM is robust to various protein types and sizes, and can be applied to more general types of mutational effects such as sequence insertions or deletions.