Improved protein structure prediction using predicted inter-residue orientations
Improved protein structure prediction using predicted inter-residue orientations
Abstract The prediction of inter-residue contacts and distances from co-evolutionary data using deep learning has considerably advanced protein structure prediction. Here we build on these advances by developing a deep residual network for predicting inter-residue orientations in addition to distances, and a Rosetta constrained energy minimization protocol for rapidly and accurately generating structure models guided by these restraints. In benchmark tests on CASP13 and CAMEO derived sets, the method outperforms all previously described structure prediction methods. Although trained entirely on native proteins, the network consistently assigns higher probability to de novo designed proteins, identifying the key fold determining residues and providing an independent quantitative measure of the “ideality” of a protein structure. The method promises to be useful for a broad range of protein structure prediction and design problems.
Ovchinnikov Sergey、Park Hahnbeom、Yang Jianyi、Anishchenko Ivan、Baker David、Peng Zhenling
John Harvard Distinguished Science Fellowship Program, Harvard UniversityDepartment of Biochemistry, University of Washington||Institute for Protein Design, University of WashingtonSchool of Mathematical Sciences, Nankai UniversityDepartment of Biochemistry, University of Washington||Institute for Protein Design, University of WashingtonDepartment of Biochemistry, University of Washington||Institute for Protein Design, University of Washington||Howard Hughes Medical Institute, University of WashingtonCenter for Applied Mathematics, Tianjin University
生物科学研究方法、生物科学研究技术生物工程学分子生物学
Ovchinnikov Sergey,Park Hahnbeom,Yang Jianyi,Anishchenko Ivan,Baker David,Peng Zhenling.Improved protein structure prediction using predicted inter-residue orientations[EB/OL].(2025-03-28)[2025-05-10].https://www.biorxiv.org/content/10.1101/846279.点此复制
评论