ADDRESSING MACROMOLECULAR CRYSTALLOGRAPHY CCD DETECTOR SPATIAL VARIATION

这个子项目是许多研究子项目中的一个 由NIH/NCRR资助的中心赠款提供的资源。子项目和 研究者（PI）可能从另一个NIH来源获得了主要资金， 因此可以在其他CRISP条目中表示。所列机构为 研究中心，而研究中心不一定是研究者所在的机构。 在高质量的大分子晶体学数据集中的对称性等效低分辨率反射通常缩放到2- 4%的R-合并值，这取决于数据冗余。然而，模拟考虑到随机和系统误差所产生的典型的实验装置预测较低的R-合并。模拟中未考虑的一个可能的误差源是检测器响应中的像素间变化。在SSRL光束线处使用CCD检测器的情况下，可能会出现这种变化，因为每根光纤具有不同的效率，但点扩散函数使它们在泛场中全部模糊。然而，尖锐的衍射斑点仅在给定的一组纤维上平均。为了检验这一假设，从同一晶体在两个不同的检测器位置收集了一个非常冗余的部分数据集。当合并来自探测器的两个不同部分的图像时，R-合并福尔斯作为冗余的平方根下降，如对于随机噪声所预期的。 然而，当分别处理来自不同探测器位置的数据集时，2.5？误差的大小与光斑大小相关，这进一步表明检测器的像素到像素的变化（即，当在较大数目的像素上平均时，由该效应引入的误差减小）。计划进行更多的实验来确定这种效应对SAD和MAD实验中异常信号测量的影响，以形成软件开发的基础，从而减轻任何影响。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Symmetry equivalent low resolution reflections in a good quality macromolecular crystallography data set typically scale to an R-merge value of 2-4 % depending on data redundancy. However, simulations taking into account random and systematic errors arising from typical experimental set-ups predict a lower R-merge. A possible source of error not accounted for in simulations is pixel-to-pixel variation in the response of the detector. In the case of the CCD detectors in use at SSRL beam lines, such variation may arise because each optical fiber has a different efficiency but the point spread function blurs them all together in the flood field. However, sharp diffraction spots only average over a given set of fibers. To test this hypothesis, a very redundant partial data set was collected from the same crystal at two different detector positions. When merging the images from two different part of the detector the R-merge falls as the square root of the redundancy, as expected for random noise. However, when processing the data sets from different detector position separately, a systematic difference of 2.5 ? 5 % was found. The size of the error correlates with the spot size, which further suggests a pixel to pixel variation for the detector (i.e., when the error introduced by this effect decreases when averaged over a larger number of pixels). More experiments to determine the influence of this effect on the measurement of the anomalous signal in SAD and MAD experiments are planned to form the basis for software developments to mitigate any effects.