Estimating the number of molecules in the crystallographic asymmetric unit is one of the first steps in a macromolecular structure determination. Based on a survey of 15,641 crystallographic Protein Data Bank (PDB) entries the distribution of V M , the crystal volume per unit of protein molecular weight, known as Matthews coefficient, has been reanalyzed. The range of values and frequencies has changed in the 30 years since Matthews first analysis of protein crystal solvent content. In the statistical analysis, complexes of proteins and nucleic acids have been treated as a separate group. In addition, the V M distribution for nucleic acid crystals has been examined for the first time. Observing that resolution is a significant discriminator of V M , an improved estimator for the probabilities of the number of molecules in the crystallographic asymmetric unit has been implemented, using resolution as additional information. , that the fraction of the crystal volume occupied by solvent ranged from 27% to 78%, with the most common value being about 43% (Matthews 1968(Matthews , 1976. Matthews defined V M , known as the Matthews coefficient, as the crystal volume per unit of protein molecular weight, and showed that V M bears a straightforward relationship to the fractional volume of solvent in the crystal. The range of V M values was found to be essentially independent of the volume of the asymmetric unit. The frequency distribution of V M for proteins is not symmetric, but has a rather sharp cutoff at the lower end, at approximately the value for close packed spheres (∼26% solvent content). Matthews recognized that the distribution of V M would be useful in preliminary studies of protein crystals to estimate the number of molecules per asymmetric unit, particularly in the molecular weight region below 70 kD, although he suggested that examples would likely be found with V M lying outside the range. Although it was also noted that higher molecular weight proteins had a tendency to form crystals with a higher fractional volume of solvent (Matthews 1976), there were not enough data to statistically determine the range of V M for such proteins.More than 30 years have passed since Matthews first analysis of protein crystal solvent content, yet the original distribution of V M is still widely used as a guide in determining the contents of the crystallographic asymmetric unit. Given the plethora of crystal forms now available in the Protein Data Bank (PDB; Berman et al. 2000), we decided

The crystal structure of the diphtheria toxin dimer at 2.5 A resolution reveals a Y-shaped molecule of three domains. The catalytic domain, called fragment A, is of the alpha + beta type. Fragment B actually consists of two domains. The transmembrane domain consists of nine alpha-helices, two pairs of which are unusually apolar and may participate in pH-triggered membrane insertion and translocation. The receptor-binding domain is a flattened beta-barrel with a jelly-roll-like topology. Three distinct functions of the toxin, each carried out by a separate structural domain, can be useful in designing chimaeric proteins, such as immunotoxins, in which the receptor-binding domain is substituted with antibodies to target other cell types.

A prototype example of a screen design and efficiency estimator program, CrysPred, is available at http://www-structure.llnl.gov/cryspred/

On average, the mother liquor or solvent and its constituents occupy about 50% of a macromolecular crystal. Ordered as well as disordered solvent components need to be accurately accounted for in modelling and refinement, often with considerable complexity.