Validated Conformational Ensembles Are Key for the Successful Prediction of Protein Complexes

详细信息    查看全文

• 作者：Carles Pons ; R. Bryn Fenwick ; Santiago Esteban-Mart铆n ; Xavier Salvatella ; Juan Fernandez-Recio
• 刊名：Journal of Chemical Theory and Computation
• 出版年：2013
• 出版时间：March 12, 2013
• 年：2013
• 卷：9
• 期：3
• 页码：1830-1837
• 全文大小：345K
• 年卷期：v.9,no.3(March 12, 2013)
• ISSN：1549-9626

文摘

Conformational fluctuations in proteins play key roles in their functions and interactions. In this work, validated conformational ensembles for ubiquitin have been used in docking trials. The ensembles were used in a systematic predictive study of known ubiquitin complexes by applying a cross-docking strategy against the bound structure of each partner. The global docking predictions obtained with the complete ubiquitin ensembles were significantly better than those obtained with the crystallographic structure of free ubiquitin. Importantly, in all cases we identified an individual ensemble member that performed equally well, or even better, than the bound structure of ubiquitin. These results unequivocally demonstrate that, for proteins that recognize binding partners by conformational selection, the availability of conformational ensembles can greatly improve the performance of automatic docking predictions. Our results highlight the need for docking methodologies to capitalize on validated ensemble representations of biomacromolecules.