Crystallographic Methods

Estimation of errors in protein structure determination.

Many biological questions associated with enzyme activity and protein transport require the determination of atomic positions of known precision. With the advent of high-throughput crystallography we also need methods of automatically detecting large systematic errors in protein structures. During our investigation of the Rfree ratio as a method of detecting errors, we became more aware of the need to be able to weight the X-ray observations relative to the geometrical restraints during protein refinement. To address this problem, we have developed a marginal likelihood method to estimate the variances of the observations and restraints directly from the observations themselves. The geometry variances should agree with the dispersion of geometrical parameters in small molecule structures provided that there are no systematic errors in the molecular model. This latter comparison provides a means of checking the model for sytematic errors. See ref (iv) for a description of this work.

References

(i) Error estimates of protein structure coordinates and deviations from standard geometry by full matrix refinement of gamma B- and beta B2-crystallin. Tickle I J, Laskowski R A & Moss D S (1998a). Acta Cryst. , D54, 243-252.

(ii) Rfree and the Rfree ratio. Part I. Derivation of expected values of cross-validation residuals used in macromolecular least-squares refinement. Tickle I J, Laskowski R A & Moss D S (1998b). Acta Cryst., D54, 547-557.

(iii) Rfree and the Rfree ratio. Part II. Calculation of the expected values and variances of cross-validation residuals in macromolecular least-squares refinement. Tickle I J, Laskowski R A & Moss D S (2000). Acta Cryst.,D56, 442-450.

(iv) Estimation of weights and validation: a marginal likelihood approach. Lebedev, A A, Tickle, I J, Laskowski, R A and Moss, D S, Acta Cryst.,(2003), D59, 1557-1566.

(v) Diffuse X-Ray scattering from molecular crystals, Moss D S, Harris G W, Wostrack A and Sansom C, Crystallography Reviews, (2003), 9, 229-277.

Any questions and bug reports to David Moss

Last updated 25 March 2008

d.moss@bbk.ac.uk