Browsing by Author "Patel, J"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Towards real-time analysis of liquid jet alignment in serial femtosecond crystallography
    (International Union of Crystallography, 2022-06-02) Patel, J; Round, A; Bielecki, J; Doerner, K; Kirkwood, HJ; Letrun, R; Schulz, J; Sikorski, M; Valiki, M; de Wijn, R; Peele, AG; Mancuso, AP; Abbey, B
    Liquid sample delivery systems are used extensively for serial femtosecond crystallography at X-ray free-electron lasers (XFELs). However, misalignment of the liquid jet and the XFEL beam leads to the X-rays either partially or completely missing the sample, resulting in sample wastage and a loss of experiment time. Implemented here is an algorithm to analyse optical images using machine vision to determine whether there is overlap of the X-ray beam and liquid jet. The long-term goal is to use the output from this algorithm to implement an automated feedback mechanism to maintain constant alignment of the X-ray beam and liquid jet. The key elements of this jet alignment algorithm are discussed and its performance is characterized by comparing the results with a manual analysis of the optical image data. The success rate of the algorithm for correctly identifying hits is quantified via a similarity metric, the Dice coefficient. In total four different nozzle designs were used in this study, yielding an overall Dice coefficient of 0.98. © 2022 The Authors, CC BY 4.0 licence
  • Thumbnail Image
    Item
    Towards real-time analysis of liquid jet alignment in SFX
    (Australian Nuclear Science and Technology Organisation, 2021-11-25) Patel, J; Peele, AG; Abbey, B; Round, AP; Mancuso, A
    Serial femtosecond crystallography (SFX) enables atomic scale imaging of protein structures via X-ray diffraction measurements from large numbers of small crystals intersecting intense X-ray Free Electron Laser (XFEL) pulses. Sample injection typically involves continuous delivery of crystals to the pulsed XFEL beam via a liquid jet. Due to movement of the jet, which is often focused to further reduce its diameter using a gas virtual dynamic nozzle (GVDN), jet position is often adjusted multiple times during the experiment. This can result in loss of beamtime and significant manual intervention. Here we present a novel approach to the problem of liquid jet misalignment in SFX based on machine vision. We demonstrate automatic identification and classification of when there is overlap (‘hit’) and when there is not overlap (‘miss’) between the XFEL beam and jet. Our algorithm takes as its input optical images from the ‘side microscope’ located inside the X-ray hutch. This algorithm will be incorporated into the control system at the SFX/SPB beamline at the European XFEL where it will be used for in-situ ‘alignment correction’ via a continuous feedback loop with the stepper motors controlling the location of the nozzle within the chamber. Full automation of this process will result in a larger volume of useful data being collected. By increasing the efficiency and reducing the per experiment operational cost of SFX at the European XFEL a higher volume of experiments can be performed. In addition, via analysis of the feedback metrology we anticipate that optimised nozzle designs and jetting conditions could be achieved further benefitting the end user.
  • Thumbnail Image
    Item
    Towards real-time analysis of liquid jet alignment in SFX
    (International Union of Crystallography, 2021-08-14) Patel, J; Round, A; Peele, AG; Mancuso, A; Abbey, B
    Serial femtosecond crystallography (SFX) enables the retrieval of the molecular structure of protein molecules at the atomic level through the measurement of large numbers of small crystals intersecting intense X-ray pulses. The method of sample delivery for SFX has a very significant impact on the success (or otherwise) of the experiment since this can impact the signal-to-noise, resolution, and amount of data that can be obtained. In particular, highly efficient sample delivery is critical, since this minimises the amount of X-ray Free Electron Laser (XFEL) beamtime required as well as reducing sample consumption and data volumes. Here we present the results from a series of liquid jet experiments performed at the European XFEL using gas focused liquid injectors, gas virtual dynamic nozzle (GVDN), and double flow focusing nozzles. Although these methods are well-established and used extensively at the European XFEL a major drawback of using these injectors is that over time the jet can become misaligned with the XFEL beam. At present, this requires regular manual monitoring in order to ensure that the relative drift of the jet with respect to the X-ray beam does not become so significant that the beam either ‘clips’ or misses the jet entirely. Manual adjustment of the liquid jet to ensure alignment with the Xray beam costs the beamline staff time, is prone to errors, and ultimately reduces the amount of useable data that is collected. In order to address the issue of jet misalignment we present a novel approach to analysing the liquid stream both with (‘hit’) and without (‘miss’) intersection by the X-ray beam using machine vision. Optical images from the from the side microscope currently used to monitor the jet are fed into our machine vision algorithm and used to classify the images as either a hit or miss. Currently we are testing the efficacy of the algorithm with a variety of nozzles and jetting conditions. The algorithm will then be incorporated into the control system at the SFX/SPB beamline at the European XFEL where it will be used to generate an ‘alignment correction’ to the stepper motors controlling the location of the nozzle within the chamber. Via a continuous feedback loop, fine adjustments will be made to the position of the liquid jet ensuring that maximum X-ray beam/liquid jet overlap is achieved. Since this process is fully automated we anticipate that it will result in a larger volume of useful data being collected without requiring any manual intervention. By increasing the efficiency and reducing the per experiment operational cost of SFX at the European XFEL ultimately more experiments can be performed. In addition, via analysis of the feedback metrology we anticipate that optimised nozzle designs and jetting conditions could be achieved further benefitting the end user. © The Authors