The energy-based refinement of low-resolution protein structure models to atomic-level accuracy is a major challenge for computational structural biology. Here we describe a new approach to refining protein structure models that focuses sampling in regions most likely to contain errors while allowing the whole structure to relax in a physically realistic all-atom force field. In applications to models produced using nuclear magnetic resonance data and to comparative models based on distant structural homologues, the method can significantly improve the accuracy of the structures in terms of both the backbone conformations and the placement of core side chains. Furthermore, the resulting models satisfy a particularly stringent test: they provide significantly better solutions to the X-ray crystallographic phase problem in molecular replacement trials. Finally, we show that all-atom refinement can produce de novo protein structure predictions that reach the high accuracy required for molecular replacement without any experimental phase information and in the absence of templates suitable for molecular replacement from the Protein Data Bank. These results suggest that the combination of high-resolution structure prediction with state-of-the-art phasing tools may be unexpectedly powerful in phasing crystallographic data for which molecular replacement is hindered by the absence of sufficiently accurate previous models.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
We thank Rosetta@home participants for contributing computing power that made testing of many new ideas possible; the DOE INCITE program for access to Blue Gene/L at Argonne National Laboratory and the IBM Blue Gene Watson supercomputers; and the NCSA, SDSC and Argonne National Laboratory supercomputer centres for computer time and help with porting Rosetta to Blue Gene. We thank D. Kim and K. Laidig for developing the computational infrastructure underlying Rosetta@home; J. Abendroth for help with RESOLVE and ARP/wARP software; M. Kennedy of NESG for the NMR structure coordinates of protein 1xpw and for help with the molecular replacement calculations; and J. Abendroth, J. Bosch, J. Havranek and C. Wang for comments on the manuscript. We also thank the CASP organizers and contributing structural biologists for providing an invaluable test set for new structure refinement methods. This work was funded by the National Institute of General Medical Sciences, National Institutes of Health (to D.B.), the Wellcome Trust, UK (to R.J.R.), the Howard Hughes Medical Institute (D.B.), a Leukemia and Lymphoma Society Career Development fellowship (to B.Q.), and a Jane Coffin Childs fellowship (to R.D.). Rosetta software and source code are available to academic users free of charge at http://www.rosettacommons.org/software/.
Author Contributions B.Q., S.R. and R.D. contributed equally to this work. Structure predictions for NMR-based, comparative-model-based and de novo predictions were carried out by S.R., B.Q. and R.D. respectively, with advice and software from D.B. and P.B. Phasing trials were performed by R.J.R., B.Q., S.R. and R.D., with advice from R.J.R. and A.J.M. All authors discussed results and commented on the manuscript.
The file contains Supplementary Tables S1-S2 and Supplementary Figures S1-S5 with Legends and additional acknowledgements.
About this article
Integrating Bonded and Nonbonded Potentials in the Knowledge-Based Scoring Function for Protein Structure Prediction
Journal of Chemical Information and Modeling (2019)