Research On Engineering Principles Of X-ray Diffraction Data Processing Of Protein Monocrystal And Software Development

As one of the most important components of organisms,protein plays an extremely important role in various life processes.To perform normal functions,proteins often need to interact with small molecules or other proteins.The study of the life process is mainly associated with the study of protein including protein structure and function,and as is known to all,structure determines function,so our study of protein structure is particularly important.Structural biologists mainly use series of methods including physical methods,biochemical and molecular biological methods to investigate the structure and function of biological macromolecules.Various methods including X-ray crystallography,nuclear magnetic resonance(NMR)and cryo-electron microscopy(Cryo-EM)could be important tools to resolve protein structures of different origins.Among them,crystal X-ray diffraction is one of the most important techniques,hence,processing of crystal X-ray diffraction data is an important part of resolving protein structure.Present study was performed to investigate the principles involved in the processing of X-ray diffraction data of biological macromolecular crystals,design and complete a set of self-developed software packages named autoPX.The autoPX package is mainly used to process the monochromatic diffraction data of bio-macromolecular crystals collected at Shanghai Synchrotron Radiation Facility(SSRF).During processing of X-ray diffraction datasets,autoPX can automatically identify the hardware and parameter information related to the diffraction experiment,and convert the original diffraction picture into the corresponding two-dimensional array.Combining the modified canny operator;(an edge recognition algorithm in digital images);with the connected component labeling algorithm,autoPX develops a new diffraction spot recognition method,which provides a guarantee for the indexing.For the inconsistency between the position of the X-ray incident light center and the diffraction data header file records caused by instrument maintenance in some beam lines of SSRF,autoPX has developed a set of algorithms for calculating and correcting the position of the incident light center utilizing the difference vector Fourier transform algorithm,which greatly improves the success rate of indexing of the diffraction data with poor quality.Based on the crystal orientation matrix obtained by indexing,autoPX can predict and correct position of the whole space index in the resolution range.With those information,autoPX can perform the simple summation integration and 2D profile fitting as well as system error correction by scaling.Refers to the format of HKL2000 result files,autoPX output files could ensure that the subsequent structure analysis-related software is recognizable.Meanwhile,it could also write the intermediate process into the log file for viewing.Considering the user feasibility,we also developed a Graphical User Interface(GUI)version for internal testing and a command-line version for domestic peers.Several diffraction datasets,including anomalous scattering,collected at SSRF have already been processed by autoPX,HKL2000(or DIALS)and XDS,respectively,followed by utilizing the subsequent software packages for phase solving,initial model building,and structural model refining.Taken together,comparative analysis of these software packages demonstrated similar quality of crystallographic data that are suitable for subsequent structural analysis,suggesting that autoPX software suite is able to be a conventional supporting software of SSRF beamlines.In addition to being able to process the diffraction data currently collected in China,we can also use this as a basis to meet the challenges brought about by new technologies.