An empirical method for correcting rotation-camera data for absorption and decay effects

When the same set of structure factors is measured twice from separate crystals, some of the differences between the two sets of data will be due to differences in absorption and crystal disorder. This will particularly affect the isomorphous-replacement method. By carefully choosing crystals of the same shape and size, and by recording the reflections in the same order, these sources of error may sometimes be minimized. However, there will be cases where serious errors of this kind will prove unavoidable. A method of deriving a correction for the absorption and decay differences between two sets of structure factors after the measurements have been made is presented. The method is designed to correct protein data collected on a rotation camera. Where an absorption-corrected subset of the observed data is available, the rotation-camera data may be corrected to this. Where this cannot be done, the method allows a relative correction between native and isomorphous-derivative data sets. The method has been implemented on a mini-computer. Results are presented which show that in some cases a substantial improvement in the quality of the data is obtained, as judged by the agreement of symmetry-related amplitudes. Furthermore, the method is capable of improving the reliability of measurements of anomalous scattering from heavy-atom-substituted isomorphous derivatives. The potential usefulness of this method is discussed.