Figure 1
Cumulative completeness analysis to determine optimal segmentation for `early' and `late' data sets as performed by autoPROC. (a) PDB entry 5srx (Gahbauer et al., 2023): the smoothed cumulative completeness as a function of increasing image number is given in red, while the associated anomalous completeness is given in blue; the related values as a function of decreasing image number are given in green and grey, respectively. The optimal image ranges to achieve high completeness, while at the same time maintaining a large distance (in image space), for the `early' and `late' data sets are marked by yellow and green backgrounds, respectively. (b) PDB entry 7kds (Seattle Structural Genomics Center for Infectious Disease, unpublished work): a high-symmetry space group (P4132) together with a large rotation range during data collection (180°) achieves complete `early' and `late' data sets separated by a large image range (and therefore dose). (c) PDB entry 7wcj (Sharma et al., 2022): the choice of an unfortunate starting angle for data collection shows as a plateauing of cumulative completeness at around image 50 (where a mirror plane in reciprocal space is crossed), leading to a slower increase of cumulative completeness while still accumulating dose. |