Z-score to evaluate the absolute quality of macromolecular models based on SAXS data

Abstract

Small-angle X-ray scattering (SAXS) is a highly popular technique to assess the native three-dimensional structure of biological macromolecules in solution. Here we introduce a statistical criterion, Z-score, as a novel quality measure of SAXS-based structural models which positively correlates with data quality. We propose that, besides a goodness-of-fit (GOF) measure such as reduced {chi}2, the Z-score reflecting the ability of a given SAXS curve to differentiate between possible models should always be reported.