![]() Instead, an iterative search is performed in which a limited number of parameters are evaluated, and new conditions are selected based on the results. Testing all combinations is impractical because the number grows significantly with the inclusion of each parameter. Evaluating each combination of parameters involves comprehensive sampling of one or more grids using a cryo-EM. These artifacts can severely limit the quality of specimens and are typically addressed through an optimization process in which several parameters are varied to increase the stability and mono-dispersity of the target macromolecule ( Passmore and Russo, 2016). In addition, vitrification methods typically yield variations in ice thickness across the grid. During specimen preparation, interactions with the air-water interface facilitated by the confinement into a thin layer of buffer can destabilize protein complexes leading to denaturation and aggregation or force the molecules into a ‘preferred orientation’ ( Noble et al., 2018). The ideal specimen for solving a structure is a single layer of randomly oriented macromolecular complexes embedded into a thin slab of vitreous ice. However, optimizing specimens for high-resolution cryo-EM imaging remains a significant barrier ( Weissenberger et al., 2021). Over the past decade, advances in hardware and software have improved the resolution and throughput of single particle analysis (SPA), establishing cryo-EM as a method of choice in structural biology. Our automated tool for systematic evaluation of specimens streamlines structure determination and lowers the barrier of adoption for cryo-EM. Manual annotations can be used to re-train the feature recognition models, leading to improvements in performance. A web interface provides remote control over the automated operation of the microscope in real time and access to images and annotation tools. SmartScope employs deep-learning-based object detection to identify and classify features suitable for imaging, allowing it to perform thorough specimen screening in a fully automated manner. Here, we present SmartScope, the first framework to streamline, standardize, and automate specimen evaluation in cryo-EM. While automation has significantly increased the speed of data collection, specimens are still screened manually, a laborious and subjective task that often determines the success of a project. EMD-25764įinding the conditions to stabilize a macromolecular target for imaging remains the most critical barrier to determining its structure by cryo-electron microscopy (cryo-EM). human DNA polymerase gamma accessory subunit, POLG2. Riccio AA, Bouvette J, Borgnia MJ, Copeland WC. Automated systematic evaluation of cryo-EM specimens with SmartScope - Trained models for square and hole detectors. īouvette J, Huang Q, Bartesaghi A, Borgnia MJ. Automated systematic evaluation of cryo-EM specimens with SmartScope - Training data for hole detector. Automated systematic evaluation of cryo-EM specimens with SmartScope - Training data for square detector. Cryo-EM density maps of POLG2 collected using SmartScope have been deposited in the Electron Microscopy Data Bank (EMDB) with accession code EMD-25764.īouvette J, Huang Q, Bartesaghi A, Borgnia MJ. ![]() Source code and installation instructions are available from (copy archived at swh:1:rev:9e58e2a2b278ca65156390175d393819fbb16a3b). Square and hole images and corresponding labels used for training the ML models are available at 10.5281/zenodo.6814642 (square finder) and 10.5281/zenodo.6814652 (hole finder). ![]() Cryo-EM density maps of POLG2 collected using SmartScope have been deposited in the Electron Microscopy Data Bank (EMDB) with accession code EMD-25764. The Jupyter notebook used to aggregate the statistic in Figure 3 and Appendix 1-Tables 2–5 is part of the code repository ( Bouvette et al., 2022a). Trained models used to obtain the results shown in Figures 1 and and2 2 are available to download from 10.5281/zenodo.6842025.
0 Comments
Leave a Reply. |