research papers 1168 Walsh et al.  Ultrafast protein structure determination Acta Cryst. (1999). D55, 1168–1173 Acta Crystallographica Section D Biological Crystallography ISSN 0907-4449 Taking MAD to the extreme: ultrafast protein structure determination Martin A. Walsh, a Irene Dementieva, a Gwyndaf Evans, a ² Ruslan Sanishvili a and Andrzej Joachimiak a,b * a Building 202, Argonne National Laboratory, 9700 South Cass Avenue, Argonne IL 60439, USA, and b Northwestern University, Depart- ment of Biochemistry, Molecular Biology and Cell Biology, Evanston IL 60208, USA ² Present address: MRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 2QH, England. Correspondence e-mail: andrzejj@anl.gov # 1999 International Union of Crystallography Printed in Denmark – all rights reserved Multiwavelength anomalous diffraction data were measured in 23 min from a 16 kDa selenomethionyl-substituted protein, producing experimental phases to 2.25 A ˚ resolution. The data were collected on a mosaic 3  3 charge-coupled device using undulator radiation from the Structural Biology Center 19ID beamline at the Argonne National Laboratory’s Advanced Photon Source. The phases were independently obtained semiautomatically by two crystallographic program suites, CCP4 and CNS. The quality and speed of this data acquisition exemplify the opportunities at third-generation synchrotron sources for high-throughput protein crystal structure deter- mination. Received 25 September 1998 Accepted 3 March 1999 PDB Reference: chaperonin apical domain, 1srv. 1. Introduction As of September 1998, 7579 protein coordinate sets have been deposited with the Protein Data Bank (Bernstein et al., 1977). Although this is a remarkable achievement, the number of structures determined pales in comparison with the number of proteins which are still being discovered. Many factors have contributed to this disparity, such as the availability of stan- dardized molecular-biology procedures and protocols in protein purification and, in particular, the progress being made in genomic analyses, the most ambitious of which is the sequencing of the human genome (Collins & Galas, 1993; Rowen et al., 1997). In spite of the tremendous advances in the structural biology field, rapid protein structure determination is still hindered by two main stumbling blocks: protein crys- tallization and initial phase determination. In order to bridge the gap between genomics and structural biology, these obstacles need to be urgently addressed. Crystallization has seen significant developments in the past decade, the most noteworthy being the use of incomplete factorial analysis of conditions which may produce suitable crystals for X-ray diffraction experiments (Carter, 1992; Jancarik & Kim, 1991). Developments in phasing techniques have also received significant attention, with much effort being put into the improvement of phases obtained from multiple isomorphous replacement, which remains the principal method for ab initio phasing of protein structures. (For a recent review, see Ke, 1997, and references therein.) Here, we concentrate on recent developments which can considerably reduce the time and effort required to obtain phase information using the multiwavelength anomalous diffraction (MAD) method, which has become popular in the last several years owing to the work by Hendrickson and co- workers (Hendrickson, 1991). The MAD method exploits resonance or anomalous scattering, which occurs when the energy of the incident X-rays approaches that of an electronic transition in an atom. The effect is energy-dependent implying