Article | Published:

cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination

Nature Methods volume 14, pages 290296 (2017) | Download Citation

Abstract

Single-particle electron cryomicroscopy (cryo-EM) is a powerful method for determining the structures of biological macromolecules. With automated microscopes, cryo-EM data can often be obtained in a few days. However, processing cryo-EM image data to reveal heterogeneity in the protein structure and to refine 3D maps to high resolution frequently becomes a severe bottleneck, requiring expert intervention, prior structural knowledge, and weeks of calculations on expensive computer clusters. Here we show that stochastic gradient descent (SGD) and branch-and-bound maximum likelihood optimization algorithms permit the major steps in cryo-EM structure determination to be performed in hours or minutes on an inexpensive desktop computer. Furthermore, SGD with Bayesian marginalization allows ab initio 3D classification, enabling automated analysis and discovery of unexpected structures without bias from a reference map. These algorithms are combined in a user-friendly computer program named cryoSPARC (http://www.cryosparc.com).

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Accessions

References

  1. 1.

    Biochemistry. The resolution revolution. Science 343, 1443–1444 (2014).

  2. 2.

    & Structural biology. Beyond blob-ology. Science 345, 617–619 (2014).

  3. 3.

    , , & Structure of the TRPV1 ion channel determined by electron cryo-microscopy. Nature 504, 107–112 (2013).

  4. 4.

    , , & Ribosome structures to near-atomic resolution from thirty thousand cryo-EM particles. eLife 2, e00461 (2013).

  5. 5.

    et al. Structure of a yeast spliceosome at 3.6-angstrom resolution. Science 349, 1182–1191 (2015).

  6. 6.

    et al. 2.3 Å resolution cryo-EM structure of human p97 and mechanism of allosteric inhibition. Science 351, 871–875 (2016).

  7. 7.

    et al. The 3.8 Å resolution cryo-EM structure of Zika virus. Science 352, 467–470 (2016).

  8. 8.

    , , , & Ensemble cryo-EM uncovers inchworm-like translocation of a viral IRES through the ribosome. eLife 5, e14874 (2016).

  9. 9.

    , , & Accelerated cryo-EM structure determination with parallelisation using GPUs in RELION-2. eLife 5, e18722 (2016).

  10. 10.

    Avoiding the pitfalls of single particle cryo-electron microscopy: Einstein from noise. Proc. Natl. Acad. Sci. USA 110, 18037–18041 (2013).

  11. 11.

    et al. Outcome of the first electron microscopy validation task force meeting. Structure 20, 205–214 (2012).

  12. 12.

    A Bayesian view on cryo-EM structure determination. J. Mol. Biol. 415, 406–418 (2012).

  13. 13.

    FREALIGN: high-resolution refinement of single particle structures. J. Struct. Biol. 157, 117–125 (2007).

  14. 14.

    Gctf: real-time CTF determination and correction. J. Struct. Biol. 193, 1–12 (2016).

  15. 15.

    , , & gEMpicker: a highly parallel GPU-accelerated particle picking tool for cryo-electron microscopy. BMC Struct. Biol. 13, 25 (2013).

  16. 16.

    Progress in digital integrated electronics. In Proc. Int. Elect. Devices Meet 11–13 (IEEE, 1975).

  17. 17.

    A maximum-likelihood approach to single-particle image refinement. J. Struct. Biol. 122, 328–339 (1998).

  18. 18.

    & Numerical Optimization (Springer, 2000).

  19. 19.

    & Optimization Models (Cambridge University Press, 2014).

  20. 20.

    Large-scale machine learning with stochastic gradient descent. In Proc. COMPSTAT'2010 (eds. Lechevallier, Y. & Saporta, G.) 177–186 (2010).

  21. 21.

    , & In Adv. Neural Inf. Process. Syst. (eds. Pereira, F., Burges, C.J.C. et al.) 1–9 (NIPS, 2012).

  22. 22.

    , , & In Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (eds. Dickinson, S. et al.) 1701–1708 (IEEE Computer Society, 2014).

  23. 23.

    , & Models for the a subunits of the Thermus thermophilus V/A-ATPase and Saccharomyces cerevisiae V-ATPase enzymes by cryo-EM and evolutionary covariance. Proc. Natl. Acad. Sci. USA 113, 3245–3250 (2016).

  24. 24.

    , & Electron cryomicroscopy observation of rotational states in a eukaryotic V-ATPase. Nature 521, 241–245 (2015).

  25. 25.

    Rigorous Global Search: Continuous Problems (Springer, 2014).

  26. 26.

    , , & An algorithm for the traveling salesman problem. Oper. Res. 11, 972–989 (1963).

  27. 27.

    , & Go-ICP: solving 3D registration efficiently and Globally optimally. In Proc. IEEE Int. Conf. Comput. Vis. (eds. Davis, L. & Hartley, R.) 1457–1464 (IEEE, 2013).

  28. 28.

    , , , & 2.8 Å resolution reconstruction of the thermoplasma Acidophilum 20S proteasome using cryo-electron microscopy. eLife 4, e06380 (2015).

  29. 29.

    et al. Cryo-EM structure of the Plasmodium falciparum 80S ribosome bound to the anti-protozoan drug emetine. eLife 3, 1–20 (2014).

  30. 30.

    & Prevention of overfitting in cryo-EM structure determination. Nat. Methods 9, 853–854 (2012).

  31. 31.

    & Optimal determination of particle orientation, absolute hand, and contrast loss in single-particle electron cryomicroscopy. J. Mol. Biol. 333, 721–745 (2003).

  32. 32.

    et al. High-resolution noise substitution to measure overfitting and validate resolution in 3D structure determination by single particle electron cryomicroscopy. Ultramicroscopy 135, 24–35 (2013).

  33. 33.

    , , , & Single particle analysis integrated with microscopy: a high-throughput approach for reconstructing icosahedral particles. J. Struct. Biol. 186, 8–18 (2014).

  34. 34.

    et al. Validation of cryo-EM structure of IP3R1 channel. Structure 21, 900–909 (2013).

  35. 35.

    , , & Three-dimensional reconstruction from a single-exposure, random conical tilt series applied to the 50S ribosomal subunit of Escherichia coli. J. Microsc. 146, 113–136 (1987).

  36. 36.

    & The orthogonal tilt reconstruction method: an approach to generating single-class volumes with no missing cone for ab initio reconstruction of asymmetric particles. J. Struct. Biol. 153, 284–299 (2006).

  37. 37.

    & Ab initio cryo-EM structure determination as a validation problem. In Proc. IEEE Int. Conf. on Image Process. (eds. Pesquet-Popescu, B. & Fowler, J.) 2090–2094 (IEEE, 2014).

  38. 38.

    et al. A statistical approach to the initial volume problem in Single Particle Analysis by Electron Microscopy. J. Struct. Biol. 189, 213–219 (2015).

  39. 39.

    , , & A Bayesian method for 3D macromolecular structure inference using class average images from single particle electron microscopy. Bioinformatics 26, 2406–2415 (2010).

  40. 40.

    & SIMPLE: Software for ab initio reconstruction of heterogeneous single-particles. J. Struct. Biol. 180, 420–427 (2012).

  41. 41.

    , & PRIME: probabilistic initial 3D model generation for single-particle cryo-electron microscopy. Structure 21, 1299–1306 (2013).

  42. 42.

    , & Building proteins in a day: Efficient 3D molecular reconstruction. In Proc. IEEE Comp. Soc. Conf. on Comput. Vis. Pattern Rec. (eds. Bischof, H. et al.) (IEEE, 2015).

  43. 43.

    , & SubspaceEM: a fast maximum-a-posteriori algorithm for cryo-EM single particle reconstruction. J. Struct. Biol. 190, 200–214 (2015).

  44. 44.

    & Low cost, high performance processing of single particle cryo-electron microscopy data in the cloud. eLife 4, e06664 (2015).

  45. 45.

    , , , & Sampling the conformational space of the catalytic subunit of human γ-secretase. eLife 4, e11182 (2015).

  46. 46.

    & Alignment of cryo-EM movies of individual particles by optimization of image translations. J. Struct. Biol. 192, 188–195 (2015).

  47. 47.

    & Measuring the optimal exposure for single particle cryo-EM using a 2.6 Å reconstruction of rotavirus VP6. eLife 4, e06980 (2015).

  48. 48.

    , , & Protocol for rapid unsupervised cryo-EM structure determination using cryoSPARC software. Protocol Exchange (2016).

  49. 49.

    , , , & EMPIAR: a public archive for raw electron microscopy image data. Nat. Methods 13, 387–388 (2016).

  50. 50.

    , , & On the importance of initialization and momentum in deep learning. J. Mach. Learn. Res. 28, 1139–1147 (2013).

Download references

Acknowledgements

We thank S. Dawood for construction of the GUI front end and members of the Rubinstein laboratory for testing cryoSPARC. A.P. was supported by a scholarship from the Natural Sciences and Engineering Research Council (NSERC), J.L.R. was supported by the Canada Research Chairs program, and D.J.F. was supported in part by the Learning in Machines and Brains program of the Canadian Institute for Advanced Research. This research was also supported by NSERC Discovery Grants (RGPIN 2015-05630 (D.J.F.) and 401724-12 (J.L.R.)) and an NVIDIA Academic Hardware Grant (M.A.B. and A.P.). Part of this work was performed while M.A.B. was a postdoctoral fellow at the University of Toronto.

Author information

Affiliations

  1. Department of Computer Science, The University of Toronto, Toronto, Ontario, Canada.

    • Ali Punjani
  2. Molecular Structure and Function Program, The Hospital for Sick Children Research Institute, Toronto, Ontario, Canada.

    • John L Rubinstein
  3. Department of Biochemistry, The University of Toronto, Toronto, Ontario, Canada.

    • John L Rubinstein
  4. Department of Medical Biophysics, The University of Toronto, Toronto, Ontario, Canada.

    • John L Rubinstein
  5. Department of Computer Science, The University of Toronto, Toronto, Ontario, Canada.

    • David J Fleet
  6. Department of Electrical Engineering and Computer Science, York University, Toronto, Ontario, Canada.

    • Marcus A Brubaker

Authors

  1. Search for Ali Punjani in:

  2. Search for John L Rubinstein in:

  3. Search for David J Fleet in:

  4. Search for Marcus A Brubaker in:

Contributions

A.P. and M.A.B. designed algorithms and implemented software. A.P., M.A.B. and J.L.R. performed experimental work. J.L.R., D.J.F., and M.A.B. contributed expertise and supervision. All authors contributed to manuscript preparation.

Competing interests

All authors are engaged in a venture to commercially support cryoSPARC for industrial use through Structura Biotechnology Inc.

Corresponding authors

Correspondence to Ali Punjani or Marcus A Brubaker.

Supplementary information

PDF files

  1. 1.

    Supplementary Text and Figures

    Supplementary Notes 1–2

  2. 2.

    Supplementary Text and Figures

    Supplementary Protocol 1

Zip files

  1. 1.

    Supplementary Data 1

    3-D Maps from CryoSPARC Refinement

  2. 2.

    Supplementary Data 2

    3-D Maps from CryoSPARC Refinement

  3. 3.

    Supplementary Data 3

    3-D Maps from CryoSPARC Refinement

  4. 4.

    Supplementary Data 4

    3-D Maps from CryoSPARC Refinement

Videos

  1. 1.

    CryoSPARC Software and Interface

    CryoSPARC (cryo-EM single particle ab initio reconstruction and classification) is a software package implementing the algorithms described in this work, along with a user-friendly web browser based interface that can be used over the internet or through local installation of the software. CryoSPARC uses graphics processing using (GPU) acceleration with self-compiling code and automatic dependency management, so the package is very simple to install. The cryoSPARC web interface can be accessed from any computer on the same network as the computer running cryoSPARC, meaning that a single desktop computer or rackmount server can provide reconstruction and classification service for an entire group of cryo-EM users who each access the software from their own computers. Remote access is also simple, for instance, using a VPN or SSH. This video shows a brief overview of the cryoSPARC web interface, demonstrating example use of the program on a real dataset, along with the dataset and experiment organization functions. Parameters are set automatically or have prepopulated defaults in the program, meaning that a user can simply select their dataset and begin a reconstruction experiment with no expertise and minimal training.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/nmeth.4169