Research ExplorerPublicationsPickYOLO: Fast deep learning particle detector for annotatio...

PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms

Standard

PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms. / Genthe, Erik; Miletic, Sean; Tekkali, Indira; Hennell James, Rory; Marlovits, Thomas C; Heuser, Philipp.

In: J STRUCT BIOL, Vol. 215, No. 3, 09.2023, p. 107990.

Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review

Harvard

Genthe, E, Miletic, S, Tekkali, I, Hennell James, R, Marlovits, TC & Heuser, P 2023, 'PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms', J STRUCT BIOL, vol. 215, no. 3, pp. 107990. https://doi.org/10.1016/j.jsb.2023.107990

APA

Genthe, E., Miletic, S., Tekkali, I., Hennell James, R., Marlovits, T. C., & Heuser, P. (2023). PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms. J STRUCT BIOL, 215(3), 107990. https://doi.org/10.1016/j.jsb.2023.107990

Vancouver

Genthe E, Miletic S, Tekkali I, Hennell James R, Marlovits TC, Heuser P. PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms. J STRUCT BIOL. 2023 Sep;215(3):107990. https://doi.org/10.1016/j.jsb.2023.107990

Bibtex

@article{a997bde8436a4024ae6c77a0e2e9166b,
title = "PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms",
abstract = "Particle localization (picking) in digital tomograms is a laborious and time-intensive step in cryogenic electron tomography (cryoET) analysis often requiring considerable user involvement, thus becoming a bottleneck for automated cryoET subtomogram averaging (STA) pipelines. In this paper, we introduce a deep learning framework called PickYOLO to tackle this problem. PickYOLO is a super-fast, universal particle detector based on the deep-learning real-time object recognition system YOLO (You Only Look Once), and tested on single particles, filamentous structures, and membrane-embedded particles. After training with the centre coordinates of a few hundred representative particles, the network automatically detects additional particles with high yield and reliability at a rate of 0.24-3.75 s per tomogram. PickYOLO can automatically detect number of particles comparable to those manually selected by experienced microscopists. This makes PickYOLO a valuable tool to substantially reduce the time and manual effort needed to analyse cryoET data for STA, greatly aiding in high-resolution cryoET structure determination.",
author = "Erik Genthe and Sean Miletic and Indira Tekkali and {Hennell James}, Rory and Marlovits, {Thomas C} and Philipp Heuser",
note = "Copyright {\textcopyright} 2023. Published by Elsevier Inc.",
year = "2023",
month = sep,
doi = "10.1016/j.jsb.2023.107990",
language = "English",
volume = "215",
pages = "107990",
journal = "J STRUCT BIOL",
issn = "1047-8477",
publisher = "Academic Press Inc.",
number = "3",

}

RIS

TY - JOUR

T1 - PickYOLO: Fast deep learning particle detector for annotation of cryo electron tomograms

AU - Genthe, Erik

AU - Miletic, Sean

AU - Tekkali, Indira

AU - Hennell James, Rory

AU - Marlovits, Thomas C

AU - Heuser, Philipp

N1 - Copyright © 2023. Published by Elsevier Inc.

PY - 2023/9

Y1 - 2023/9

N2 - Particle localization (picking) in digital tomograms is a laborious and time-intensive step in cryogenic electron tomography (cryoET) analysis often requiring considerable user involvement, thus becoming a bottleneck for automated cryoET subtomogram averaging (STA) pipelines. In this paper, we introduce a deep learning framework called PickYOLO to tackle this problem. PickYOLO is a super-fast, universal particle detector based on the deep-learning real-time object recognition system YOLO (You Only Look Once), and tested on single particles, filamentous structures, and membrane-embedded particles. After training with the centre coordinates of a few hundred representative particles, the network automatically detects additional particles with high yield and reliability at a rate of 0.24-3.75 s per tomogram. PickYOLO can automatically detect number of particles comparable to those manually selected by experienced microscopists. This makes PickYOLO a valuable tool to substantially reduce the time and manual effort needed to analyse cryoET data for STA, greatly aiding in high-resolution cryoET structure determination.

AB - Particle localization (picking) in digital tomograms is a laborious and time-intensive step in cryogenic electron tomography (cryoET) analysis often requiring considerable user involvement, thus becoming a bottleneck for automated cryoET subtomogram averaging (STA) pipelines. In this paper, we introduce a deep learning framework called PickYOLO to tackle this problem. PickYOLO is a super-fast, universal particle detector based on the deep-learning real-time object recognition system YOLO (You Only Look Once), and tested on single particles, filamentous structures, and membrane-embedded particles. After training with the centre coordinates of a few hundred representative particles, the network automatically detects additional particles with high yield and reliability at a rate of 0.24-3.75 s per tomogram. PickYOLO can automatically detect number of particles comparable to those manually selected by experienced microscopists. This makes PickYOLO a valuable tool to substantially reduce the time and manual effort needed to analyse cryoET data for STA, greatly aiding in high-resolution cryoET structure determination.

U2 - 10.1016/j.jsb.2023.107990

DO - 10.1016/j.jsb.2023.107990

M3 - SCORING: Journal article

C2 - 37364763

VL - 215

SP - 107990

JO - J STRUCT BIOL

JF - J STRUCT BIOL

SN - 1047-8477

IS - 3

ER -