Blanc, E. & Chapman, M. S. (1997). RSRef:  Interactive real-space refinement with stereochemical restraints for use during model-building. Journal of Applied Crystallography 30, 566-7.  Prepublication version of manuscript.

RSRef: Interactive real-space refinement with stereochemical restraints for use during model-building

Eric Blanc & Michael S. Chapman

Submitted (1/23/97) Computer Program Abstract to Journal of Applied Crystallography

* Department of Chemistry & Institute of Molecular Biophysics, Florida State University, Tallahassee, FL 32306-3015, USA.

The crystallographic problem

During macromolecular refinement, building the initial model, and subsequent rebuilding (Jones et al., 1991) using interactive computer graphics, are both rate-limiting and quality-determining steps. Optimizing the fit to the electron density map, while retaining good stereochemistry can be laborious.

Method of solution

Real-space refinement is used to least-squares optimize the agreement of electron density calculated from the atomic model with the experimental electron density map. Although map quality can be limiting, Chapman & Blanc (1997) showed that protein models could be improved even when refined against poor multiple isomorphous replacement (MIR) maps. The core of the algorithm was described previously (Chapman, 1995; Chapman & Rossmann, 1996) and differs from prior real-space refinements (Diamond, 1971; Jones & Liljas, 1984) with explicit accounting for the resolution and incorporation of full stereochemical restraints from TNT (Tronrud et al., 1987), both of which lead to significantly improved performance.

RSRef release 2.0 facilitates interactive modeling. Refinement options (such as refinement weights, convergence criteria, data files etc.) can now be set within a graphics user interface (GUI) (program The user selects a zone or volume to be refined within the GUI or by picking atoms within "O" (Jones et al., 1991). At the start of refinement, SelectCoord adds neighboring atoms (including crystallographic and non-crystallographic symmetry equivalents), converts formats (e.g. pdb to tnt), and writes a table of symmetry constraints, according to options selected in the GUI. Refinement cycles include RSRef (Chapman, 1995) Geometry and Shift (Tronrud et al., 1987) with ReExpand enforcing symmetry. Running summary statistics are written to the GUI. Coordinates can be reviewed in O prior to acceptance.

Solfware environment

The GUI is run by a Perl (Wall & Schwartz, 1991) program on a WWW server writing a HTML form (Graham, 1996). Parameters may be initialized from a user file (on the client computer) and are edited using an internet browser of choice. During the editing, parameters are saved in a temporary server file, with a name chosen to avoid user collisions. Parameters may be saved to a local client file or sent directly to a client Perl program, rsref_client, that: (1) runs the refinement programs; (2) parses the printed output and calculates summary statistics. Other programs are written in C and Fortran.

Hardware environment

RSRef programs have been developed on a Silicon Graphics (SG) Indigo II, and should run on any modest Unix workstation. "O" is supported on various graphics workstations (Jones et al., 1991).

Program specification

Performance statistics are given in Table 1. There are 5,000 lines in programs, 10,000 in subroutines libraries, and 7,000 in documentation. The programs have been applied to about 1 dozen different systems.


Documentation is written in HTML and can be browsed over the internet or installed locally.


RSRef is distributed under license over the internet from It is distributed without charge to academic users. Users will also need a license for TNT (Tronrud et al., 1987), and "O" (Jones et al., 1991).


This work was generously supported by the National Science Foundation (MSC; BIR9418741).


 Chapman, M. S. (1995). Restrained Real-Space Macromolecular Atomic Refinement using a New Resolution-Dependent Electron Density Function. Acta Crystallographica A51, 69-80.
Chapman, M. S. & Blanc, E. (1997). Potential use of Real Space Refinement in Protein Structure Determination. Acta Crystallographica D53, 203-6.
Chapman, M. S. & Rossmann, M. G. (1996). Structural Refinement of the DNA-containing Capsid of Canine Parvovirus using RSRef, a Resolution-Dependent Stereochemically Restrained Real-Space Refinement Method. Acta Crystallographica D52, 129-42.
Diamond, R. (1971). A Real-Space Refinement Procedure for Proteins. Acta Crystallographica A27, 436-452.
Graham, I. S. (1996). HTML Sourcebook. 2nd edit, Wiley, New York.
Jones, T. A. & Liljas, L. (1984). Crystallographic Refinement of Macromolecules having Non-crystallographic Symmetry. Acta Crystallographica A 40, 50-7.
Jones, T. A., Zou, J.-Y., Cowan, S. W. & Kjeldgaard, M. (1991). Improved Methods for Building Protein Models in Electron Density Maps and the Location of Errors in these Models. Acta Crystallographica A47, 110-9.
Tronrud, D. E., Ten Eyck, L. F. & Matthews, B. W. (1987). An Efficient General-Purpose Least-Squares Refinement Program for Macromolecular Structures. Acta Crystallographica A43, 489-501.
Wall, L. & Schwartz, R. L. (1991). Programming perl, O'Reilly & Associates, Inc., Sebastapol, CA.

Selected refinement region
Real time
Amino acids
110 s
25 min

Table 1: Performance statistics: Refinements were at 2.0 using a 0.8 map grid and a Silicon Graphics Indigo II computer with a MIPS 4400 cpu.