The Phenix refinement framework

1: Introduction Many questions of biological significance require highly accurate knowledge of structural parameters such as atomic positions, atomic displacement parameters (ADP, also known as " B-factors ") and occupancies. The refinement of these structural parameters is therefore an essential step of macromolecular structure determination. As part of the Phenix collaboration (Adams et al., 2004) we have developed new refinement tools to increase the automation of refinement. Macromolecular structure refinement combines a large number of very diverse steps. The current implementation of the Phenix refinement protocol is shown in Figure 1. Making use of modern software development technology, each of the major building blocks is implemented as a reusable set of modules. Most of the modules are available through the open-source cctbx libraries (Grosse-Kunstleve et al., 2002; http://cctbx.sourceforge.net/) which will be included in future CCP4 releases. Some of the cctbx modules make use of CCP4 developments: the Monomer library (Vagin & Murshudov, 2004; Vagin et al., 2004) and the CMTZ library. The following sections are a brief description of the practical implementation of the Phenix refinement framework, with pointers to open-source modules that are available to the developer community. An overview of the open source libraries can be found in the series of recent IUCr Computing Commission Newsletter articles, issues 1-5 To initiate refinement, four major sources of information have to be processed: