Fu Zheng-Qing, Geisbrecht Brian V, Bouyain Samuel, Dyda Fred, Chrzas John, Kandavelu Palani, Miller Darcie J, Wang Bi-Cheng
SER-CAT, Advanced Photon Source, Argonne National Laboratory Argonne, IL 60439, USA.
Department of Biochemistry & Molecular Biology, University of Georgia Athens, GA 30602, USA.
bioRxiv. 2025 Feb 2:2024.12.10.627855. doi: 10.1101/2024.12.10.627855.
X-ray crystal diffraction has provided atomic-level structural information on biological macromolecules. Data quality determines the reliability of structural models. In most cases, multiple data sets are available from different crystals and/or collected with different experimental settings. Reliable metrics are critical to rank and select the data set with the highest quality. Many measures have been created or modified for data quality evaluation. However, some are duplicate in functionality, and some are likely misused due to misunderstanding, which causes confusion or problems, especially at synchrotron beamlines where experiments proceed quickly. In this work, these measures are studied through both theoretical analysis and experimental data with various characteristics, which demonstrated that: 1). {Rmerg, Rmeas, Rpim, CC1/2} all measure the equivalence of reflections, and the low-shell values of these metrics can be used as reliable indicators for correctness (or trueness) of Laue symmetry; 2). High-shell is a reliable and better indicator to select resolution cutoff while the overall value measures the overall strength of the data.
X射线晶体衍射已提供了关于生物大分子的原子水平结构信息。数据质量决定了结构模型的可靠性。在大多数情况下，可以从不同晶体获得多个数据集和/或使用不同实验设置进行收集。可靠的指标对于对最高质量的数据集进行排名和选择至关重要。已经创建或修改了许多用于数据质量评估的措施。然而，有些在功能上是重复的，有些可能由于误解而被误用，这会导致混乱或问题，尤其是在实验进行迅速的同步加速器光束线处。在这项工作中，通过理论分析和具有各种特征的实验数据对这些措施进行了研究，结果表明：1). {Rmerg、Rmeas、Rpim、CC1/2} 均用于测量反射的等效性，这些指标的低壳层值可作为劳厄对称性正确性（或真实性）的可靠指标；2). 高壳层是选择分辨率截止的可靠且更好的指标，而总体值衡量数据的总体强度。
