The largest impact on mainstream improvement of results was probably due to making ML (Maximum likelihood)-based 2D and 3D classification including Bayesian statistics practically feasible by clever implementation on today's computer hardware. While earlier work was indispensable to point the way, only later algorithms were optimized enough for practical use (Frealign, Relion, EMAN2, Frealing/cisTEM, Spire, and cryoSPARC).
Refines macromolecular structures by single-particle analysis of electron cryo-microscopy (cryo-EM) data. RELION employs an empirical Bayesian approach. It can be utilized for both simulated and experimental data. This tool estimates the accuracy of alignment of individual particles and the relative contribution of the different frequencies therein. It enables the correction of per-particle defocus variations.
Reconstructs single particle from electron cryo-microscopy (cryoEM) images. EMAN is capable of processing large data sets and focuses on processing data from transmission electron microscopes. The application is mainly dedicated to single particle rebuilding but includes more than 40 functionalities allowing users to perform helical reconstruction or single particle tomography as well as interacting with external software such as Frealign or Relion.
Calculates and refines three-dimensional (3D) structures of macromolecular assemblies that are calculated from images collected on an electron microscope. Frealign provides a fast and accurate projection matching algorithm, and calculates 3D reconstructions that are fully corrected for the contrast transfer function (CTF) of the microscope. Other features include refinement of microscope defocus and magnification, correction for the Ewald sphere curvature, processing of helical particles, 3D classification, and density masking. Furthermore, algorithms and run scripts were developed to take advantage of parallel computing environments to speed up processing.
Enables rapid, unbiased, and high-throughput structure discovery of proteins and molecular complexes from single-particle cryo-EM data. The algorithms underlying cryoSPARC enable high-resolution reconstructions of research and drug targets within minutes of collecting microscope data, and without the need for prior knowledge of the target structure. Using cryoSPARC can remove the risk of biased results, allow the discovery of unexpected structures, and speed up Cryo-electron microscopy workflow by orders of magnitude. Furthermore, the graphical user interface allows multiple users within a laboratory to have separate accounts, access the program remotely, upload and share data sets, manage experimental results, launch computational tasks, and view results streaming in real time as they are computed.
Allows access to cryo electron microscopy (cryo-EM) with the goal of quality assessment and result reproducibility by statistical resampling. SPHIRE is a software suited for cryo-EM novices, and users can find an array of programs which guide through the complete process.
Permits the treatment of cryo-electron microscopy (cryo-EM) images of macromolecular complexes. cisTEM is able to extract high-resolution 3D reconstructions form these images. It can deal with movies, micrographs and stacks of single-particle images, implementing a complete pipeline of processing steps. This tool is based on a maximum-likelihood algorithm. Users can define a combination of particle image stack, alignment parameters and a 3D reference structure.