DRPnet: automated particle picking in cryo-electron micrographs using deep regression.

Publisher: BioMed Central Country of Publication: England NLM ID: 100965194 Publication Model: Electronic Cited Medium: Internet ISSN: 1471-2105 (Electronic) Linking ISSN: 14712105 NLM ISO Abbreviation: BMC Bioinformatics Subsets: MEDLINE

Original Publication: [London] : BioMed Central, 2000-

Cryoelectron Microscopy* ; Electrons* ; Image Processing, Computer-Assisted* ; Regression Analysis*; Imaging, Three-Dimensional ; Neural Networks, Computer ; Proteins ; Signal-To-Noise Ratio

Background: Identification and selection of protein particles in cryo-electron micrographs is an important step in single particle analysis. In this study, we developed a deep learning-based particle picking network to automatically detect particle centers from cryoEM micrographs. This is a challenging task due to the nature of cryoEM data, having low signal-to-noise ratios with variable particle sizes, shapes, distributions, grayscale variations as well as other undesirable artifacts.
Results: We propose a double convolutional neural network (CNN) cascade for automated detection of particles in cryo-electron micrographs. This approach, entitled Deep Regression Picker Network or "DRPnet", is simple but very effective in recognizing different particle sizes, shapes, distributions and grayscale patterns corresponding to 2D views of 3D particles. Particles are detected by the first network, a fully convolutional regression network (FCRN), which maps the particle image to a continuous distance map that acts like a probability density function of particle centers. Particles identified by FCRN are further refined to reduce false particle detections by the second classification CNN. DRPnet's first CNN pretrained with only a single cryoEM dataset can be used to detect particles from different datasets without retraining. Compared to RELION template-based autopicking, DRPnet results in better particle picking performance with drastically reduced user interactions and processing time. DRPnet also outperforms the state-of-the-art particle picking networks in terms of the supervised detection evaluation metrics recall, precision, and F-measure. To further highlight quality of the picked particle sets, we compute and present additional performance metrics assessing the resulting 3D reconstructions such as number of 2D class averages, efficiency/angular coverage, Rosenthal-Henderson plots and local/global 3D reconstruction resolution.
Conclusion: DRPnet shows greatly improved time-savings to generate an initial particle dataset compared to manual picking, followed by template-based autopicking. Compared to other networks, DRPnet has equivalent or better performance. DRPnet excels on cryoEM datasets that have low contrast or clumped particles. Evaluating other performance metrics, DRPnet is useful for higher resolution 3D reconstructions with decreased particle numbers or unknown symmetry, detecting particles with better angular orientation coverage.

J Biol Chem. 2019 Mar 29;294(13):5181-5197. (PMID: 30804214)
J Comput Chem. 2004 Oct;25(13):1605-12. (PMID: 15264254)
J Mol Biol. 2003 Oct 31;333(4):721-45. (PMID: 14568533)
Elife. 2015 Mar 11;4:. (PMID: 25760083)
Ultramicroscopy. 1992 Jan;40(1):33-53. (PMID: 1580010)
Res Comput Mol Biol. 2018 Apr;10812:245-247. (PMID: 29707703)
Science. 2015 Jun 5;348(6239):1147-51. (PMID: 25953817)
Curr Opin Struct Biol. 2019 Oct;58:214-223. (PMID: 31400843)
Nat Commun. 2017 Sep 20;8(1):629. (PMID: 28931821)
Nat Methods. 2012 Sep;9(9):853-4. (PMID: 22842542)
IEEE Trans Pattern Anal Mach Intell. 2017 Jun;39(6):1137-1149. (PMID: 27295650)
J Struct Biol. 2012 Mar;177(3):630-7. (PMID: 22366277)
Nat Methods. 2014 Jan;11(1):63-5. (PMID: 24213166)
IEEE Trans Neural Netw Learn Syst. 2019 Nov;30(11):3212-3232. (PMID: 30703038)
Nature. 2018 Sep;561(7724):565-567. (PMID: 30254359)
Nature. 1984 Mar 1-7;308(5954):32-6. (PMID: 6322001)
Methods Enzymol. 2016;579:125-57. (PMID: 27572726)
IEEE Trans Pattern Anal Mach Intell. 2020 Feb;42(2):386-397. (PMID: 29994331)
J Struct Biol. 2009 May;166(2):126-32. (PMID: 19269332)
J Struct Biol. 2016 Sep;195(3):325-336. (PMID: 27424268)
Ultramicroscopy. 1987;23(1):39-51. (PMID: 3660491)
J Struct Biol. 2014 Apr;186(1):1-7. (PMID: 24607413)
Elife. 2018 Nov 09;7:. (PMID: 30412051)
Nat Struct Mol Biol. 2012 Dec;19(12):1203-7. (PMID: 23211764)
Nature. 2013 Dec 5;504(7478):107-12. (PMID: 24305160)
Methods Enzymol. 2010;482:73-100. (PMID: 20888958)
J Struct Biol. 2015 Feb;189(2):114-22. (PMID: 25486611)
Commun Biol. 2019 Jun 19;2:218. (PMID: 31240256)
Annu Rev Biophys Biophys Chem. 1986;15:237-57. (PMID: 3087377)
FEBS Lett. 2018 Oct;592(19):3229-3238. (PMID: 30184263)

Keywords: 3D reconstruction; Autopicking; Convolutional neural network; CryoEM; Deep learning; Electron microscopy; Image segmentation; Particle picking; Regression; Single particle analysis; Single particle reconstruction

0 (Proteins)

Date Created: 20210209 Date Completed: 20210217 Latest Revision: 20241007

20250114

PMC7869254

10.1186/s12859-020-03948-x

33557750