PHENIX: new methods for automation in macromolecular crystallography

University Of Calif-Lawrenc Berkeley Lab, Berkeley CA

Investigators

Linked publications, trials & patents

Abstract

The Phenix software provides methods for structure solution, refinement and validation using experimental crystallographic diffraction data. In the next five years these methods will be extended to address the solution of difficult structures that currently elude automated analysis. To make this possible research will be pursued in several complementary areas: develop methods to increase signal-to-noise, account for the effects of resolution in structure solution and refinement, integrate additional sources of information in structure solution and refinement, and develop new automation algorithms targeted at solving difficult structures. Advances in these areas will enable solution of structures that fail with current methods, and will automate and improve those that currently require highly manual effort. Low-resolution data (3.0 - 3.5 A or worse) presents one of the greatest challenges to structure solution and is currently poorly addressed by automated methods. New methods will be developed to better account for resolution throughout the structure solution and refinement process, with appropriate model parameterizations, targets, scoring functions, fragment libraries, map evaluations, and rebuilding strategies. In addition, low-resolution structures typically lack sufficient experimental data to well define the underlying structure. Additional empirical and theoretical sources of prior knowledge will be integrated into structure solution, in particular combining the power of ab-initio and structure-modeling algorithms with that of crystallographic model building and refinement. With new resolution-tuned methods and additional data, accuracy will be improved as measured by density fit, geometry, sterics, and self-consistency of each model. The application of both current and new methods will be automated and integrated such that structure solution, including the difficult cases, requires less manual interpretation. This will also allow a thorough and routine assessment of the impacts of different pathways through structure solution. The increased power of these methods will allow automation of structure determination at lower resolutions than is currently feasible, saving large amounts of time and money for structural biologists around the world in academia, government and industry, and allowing levels of error analysis and validation that are completely impractical when structure determination is performed manually.