Work

Fold-Switching Proteins Push the Boundaries of Conformational Ensemble Prediction

Year: 2026

Type: review

Abstract: A protein's function depends critically on its conformational ensemble, a collection of energy-weighted structures whose balance depends on temperature and environment. Though recent deep learning methods have substantially advanced predictions of single protein structures, computationally modeling conformational ensembles remains a challenge. Here, we focus on modeling fold-switching proteins, which remodel their secondary and/or tertiary structures and change their functions in response to cellular stimuli. These underrepresented members of the protein universe serve as test cases for a method's generalizability. They reveal that deep learning models often predict conformational ensembles by association with training set structures, which limits generalizability. These observations suggest use cases for when deep learning methods will likely succeed or fail. Developing computational methods that successfully identify new fold-switching proteins from large pools of candidates may advance modeling conformational ensembles more broadly.

Source: Annual Review of Biomedical Data Science

DOI: 10.1146/annurev-biodatasci-092524-114822

Authors: Myeongsang Lee, Lauren L. Porter

Institutions: National Institutes of Health, United States National Library of Medicine, National Heart, Lung, and Blood Institute

Language: English

Topic: Protein Structure and Dynamics

Subfield: Molecular Biology

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

Open Access status: hybrid

Funders: U.S. Department of Health and Human Services, National Institutes of Health, U.S. National Library of Medicine, NHLBI Division of Intramural Research

Awards: G1809504234, G9000017536