Improved estimation of the radius of gyration from small-angle x-ray scattering data

Small-angle X-ray scattering (SAXS) is an experimentally simple technique that provides access to low-resolution information about biological macromolecules in solution. We here provide R code and example data sets for a new algorithm that produces accurate and precise values for the radius of gyration, Rg, of a particle. Theory states that the information derived from the lowest scattering angles can be used to estimate Rg. The value Rg is a fundamental structural parameter that is related to a molecule's size and shape. The original algorithm implemented with the R code estimates Rg with a reliable variance estimate and with higher precision than the classical method. A bias-variance criterion is minimized to determine the optimal number of data points to calculate Rg. After accounting for correlation in the data, least squares regression is used to estimate the radius of gyration and an accurate variance estimate. The software also supports the use of replicate data. Use of the code and examples is described in README.pdf.