Tables for
Volume F
Crystallography of biological macromolecules
Edited by E. Arnold, D. M. Himmel and M. G. Rossmann

International Tables for Crystallography (2012). Vol. F, ch. 11.4, pp. 292-294   | 1 | 2 |

Section 11.4.12. HKL-2000 and HKL-3000

Z. Otwinowski,a* W. Minor,b D. Boreka and M. Cymborowskib

aUT Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, Dallas, TX 75390–9038, USA, and bDepartment of Molecular Physiology and Biological Physics, University of Virginia, 1300 Jefferson Park Avenue, Charlottesville, VA 22908, USA
Correspondence e-mail:

11.4.12. HKL-2000 and HKL-3000

| top | pdf |

DENZO and SCALEPACK form the numerical processing core of the HKL package. These programs can be used directly by editing commands in input scripts, but most of the time they are run through HKL-2000 and its more expanded version HKL-3000.

The basic mode of HKL-2000 reads in previously collected data as input and produces scaled and merged reflections as output. This use can be extended in three directions: control of the data-collection process by HKL-2000 and HKL-3000 (Minor et al., 2002[link], 2006[link]), incorporation of methods of structure solution by HKL-3000 (Minor et al., 2006[link]), and storage of critical intermediate results in an external database (Grabowski et al., 2007[link]).

Crystallographic structure determinations encompass a wide range of project dynamics. There are many projects where all the steps are executed serially, while others involve substantial iterative improvements, where one or a few stages are repeated a number of times. HKL-2000 and HKL-3000 are designed to make both types of project more effective. To accomplish this, information is automatically propagated between various stages of analysis, and many necessary data transformations are performed to accommodate the interface requirements of many programs and beamline controls. At the same time, crystallography requires the experimenter to be actively engaged in decision making, depending on the nature of a particular project and types of problems encountered. The experimenter may need to assess the quality of a set of crystals, decide how to collect data sets from a chosen subset, determine the symmetry of each diffraction pattern, reassess the crystal quality based on the integration and merging steps, and then solve the structure using an appropriate method. Not all programs are fully automatic, and the experimenter may need to be involved in defining how a particular procedure should be executed. As a consequence, both HKL-2000 and HKL-3000 have a multiplicity of interface screens (accessed by tabs in the graphical control centre), each of them designed to control a particular process.

The versions of HKL-2000 and HKL-3000 that interface with data-collection systems can coordinate all parameters of the diffraction experiment. This facilitates interactive experiments in which data analysis is done online and results are automatically updated when new data are collected. In such experiments, it is possible to adjust the data-collection strategy to guarantee the desired result, particularly with regard to data completeness. The strategy takes into account limitations arising from radiation damage. Radiation damage can be estimated from past experience with similar crystals, by theoretical calculations of decay based on beam intensity, and by evaluating scale- and B-factor changes in real time.

The graphical control centre of HKL-2000 (and HKL-3000) consists of three components: an internal database (optionally connected to an external one), a transition-state engine and a graphical user interface (GUI). The internal database stores all the information about data processing and data collection. It can describe not only the data already collected, but also those being collected and even those planned for collection. Each datum entered or program executed, including the data-collection interface, induces a change in the database by the transition-state engine. One of the main functions of the GUI is input to and editing of the database. The other major function is generation of reports from the database (to visualize its status).

The internal database abstraction is based on the following hierarchy: instrument type; site; experiment; crystal; three-dimensional (3D) group of diffraction images; and diffraction image. Each lower level of the hierarchy inherits the properties of the higher levels. When a program finishes analysing data at a particular level, the parent (higher-level) data are updated, so that data at the same level communicate only through the change of state of their common parent. The site-level data are created only when diffraction from a new detector is seen or when the parameters of the detector are changed, which is done rarely and typically by the X-ray equipment administrator. The experiment-level data describe diffraction data from one or more crystals of the same space group. A uniform series of diffraction images form 3D groups. There is no limit to the number of 3D groups, and, in the case of non-uniformity in the series (e.g. as found during data analysis), one 3D group can be split into two or more smaller 3D groups. The smallest 3D group can consist of one image. A crystal datum contains a set of 3D groups with a relative orientation and exposure level known a priori. In practice, this means that diffraction data encapsulated within a single crystal datum were collected from one sample at one site with potentially different settings of goniostat, data-collection axis, crystal translation, detector position, detector mode (e.g. binned/unbinned) or exposure level.

HKL-3000 combines a number of existing macromolecular crystallographic computer programs [SHELX (Sheldrick, 2008[link]), CCP4 (Collaborative Computational Project, No. 4, 1994[link]), SOLVE/RESOLVE (Terwilliger, 2004[link]), ARP/wARP (Perrakis et al., 1999[link]) and COOT (Emsley & Cowtan, 2004[link])] and decision-making algorithms into a powerful expert system (Fig.[link]). The typical end result of HKL-3000 is an interpretable electron-density map with a partially built structure and, in some cases, an almost complete and refined model.


Figure | top | pdf |

General layout of HKL-3000. The experimenter is not limited to movement between neighbouring steps. For example, using feedback from model building, one can decide to collect more data (for example, a second wavelength or a low-resolution pass) and jump directly back to the `Data collection and reduction' step.

The HKL-3000 system is designed to evaluate the results of laboratory and synchrotron diffraction experiments very quickly. The system supports most common types of macromolecular crystallography experiments: isomorphous (IR) and molecular replacement (MR), multiple anomalous diffraction (MAD)/single anomalous diffraction (SAD), and native data collection for high-resolution refinement of previously solved models of proteins or protein–ligand complexes. SAD experiments have became popular in recent years (Chruszcz et al., 2008[link]) at least partly due to development of integrated systems like HKL-3000, which allow the experimenter to validate whether experimental data collected at one wavelength are of sufficient quality to solve the structure. The concurrent data collection, data processing and quick preliminary structure solution made by HKL-3000 verifies the success of the X-ray experiment and allows optimization of the data-collection strategy while the crystal is still on the goniostat. This allows the experimenter to decide whether the experiment has been successfully completed and thus whether the crystal can be removed.


Collaborative Computational Project, Number 4 (1994). The CCP4 suite: programs for protein crystallography. Acta Cryst. D50, 760–763.
Emsley, P. & Cowtan, K. (2004). Coot: model-building tools for molecular graphics. Acta Cryst. D60, 2126–2132.
Chruszcz, M., Wlodawer, A. & Minor, W. (2008). Determination of protein structures – a series of fortunate events. Biophys. J. 95, 1–9.
Grabowski, M., Joachimiak, A., Otwinowski, Z. & Minor, W. (2007). Structural genomics: keeping up with expanding knowledge of the protein universe. Curr. Opin. Struct. Biol. 17, 347–353.
Minor, W., Cymborowski, M. & Otwinowski, Z. (2002). Automatic system for crystallographic data collection and analysis. Acta Phys. Pol. A, 101, 613–619.
Minor, W., Cymborowski, M., Otwinowski, Z. & Chruszcz, M. (2006). HKL-3000: the integration of data reduction and structure solution – from diffraction images to an initial model in minutes. Acta Cryst. D62, 859–866.
Perrakis, A., Morris, R. & Lamzin, V. S. (1999). Automated protein model building combined with iterative structure refinement. Nat. Struct. Biol. 6, 458–463.
Sheldrick, G. M. (2008). A short history of SHELX. Acta Cryst. A64, 112–122.
Terwilliger, T. (2004). SOLVE and RESOLVE: automated structure solution, density modification and model building. J. Synchrotron Rad. 11, 49–52.

to end of page
to top of page