Improved Estimation of Structure-Factor Difference Amplitudesfrom Poorly Accurate Data

The influence of measurement errors on structure-factor difference amplitudes are discussed and a formula is derived using Bayesian statistics that gives better estimates of the difference amplitudes and that reduces the noise in difference maps. The formula is of importance for reflections with poor signal-to-noise ratio. Significant improvement is obtained for poor data sets, such as those recorded in fast time-resolved experiments, and also for subsets in any data set that normally contains poorly accurate data, e.g. close to the high-resolution limit.