Summary

Protein unfolding cooperativity (whether a protein unfolds all at once or independently in different regions) is more difficult to predict than protein stability (1). Even models predicting intra-family differences could only achieve correlations of 0.16-0.24, compared to 0.40-0.53 for stability when the same input features were used. Protein language models were especially inept at predicting these.

Figures

Ref (1)

1.
Ferrari ÁJR, Dixit SM, Thibeault J, Garcia M, Houliston S, Ludwig RW, et al. Large-scale discovery, analysis, and design of protein energy landscapes. openRxiv; 2025. Available from: https://doi.org/10.1101/2025.03.20.644235