Analysing the significance of small conformational changes and low occupancy states in serial crystallographic data

The interpretation of electron density maps from time-resolved serial diffraction experiments is often hindered by incomplete initiation and mixtures of states. Additionally, it can be challenging to determine the significance of small conformational changes. Here, we present a protocol that exploits the inherent oversampling of serial crystallographic data through batch resampling and principal component analysis (PCA). This approach provides insight into the significance of small conformational changes in proteins along a reaction time course. When combined with extrapolation of structure factor amplitudes, the method further helps in the identification of low-occupancy intermediates. In this protocol report, we describe a practical workflow for batch resampling, scaling and clustering, and provide guidance for the effective use of open-source software including RoPE and Xtrapol8.