DeepFoci: Deep learning-based algorithm for fast automatic
analysis of DNA double-strand break ionizing radiation-induced
foci

J 2021

DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci

VIČAR, Tomáš, Jaromír GUMULEC, Radim KOLAR, Olga KOPECNA, Eva PAGACOVA et. al.

Základní údaje

Originální název

DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci

Autoři

VIČAR, Tomáš (203 Česká republika, domácí), Jaromír GUMULEC (203 Česká republika, domácí), Radim KOLAR (203 Česká republika), Olga KOPECNA (203 Česká republika), Eva PAGACOVA (203 Česká republika), Iva FALKOVA a Martin FALK (203 Česká republika, garant)

Vydání

Computational and Structural Biotechnology Journal, Amsterdam, Elsevier, 2021, 2001-0370

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10608 Biochemistry and molecular biology

Stát vydavatele

Nizozemské království

Utajení

není předmětem státního či obchodního tajemství

Odkazy

URL

Impakt faktor

Impact factor: 6.155

Kód RIV

RIV/00216224:14110/21:00123920

Organizační jednotka

Lékařská fakulta

DOI

http://dx.doi.org/10.1016/j.csbj.2021.11.019

UT WoS

000731411300007

Klíčová slova anglicky

DNA Damage and Repair; Ionizing Radiation-Induced Foci (IRIFs); Biodosimetry; Deep Learning; Convolutional Neural Network; Morphometry; Confocal Microscopy; Image Analysis

Štítky

14110518, rivok

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 24. 1. 2022 08:36, Mgr. Tereza Miškechová

Anotace

V originále

DNA double-strand breaks (DSBs), marked by ionizing radiation-induced (repair) foci (IRIFs), are the most serious DNA lesions and are dangerous to human health. IRIF quantification based on confocal microscopy represents the most sensitive and gold-standard method in radiation biodosimetry and allows research on DSB induction and repair at the molecular and single-cell levels. In this study, we introduce DeepFoci - a deep learning-based fully automatic method for IRIF counting and morphometric analysis. DeepFoci is designed to work with 3D multichannel data (trained for 53BP1 and gamma H2AX) and uses U-Net for nucleus segmentation and IRIF detection, together with maximally stable extremal region-based IRIF segmentation. The proposed method was trained and tested on challenging datasets consisting of mixtures of nonir-radiated and irradiated cells of different types and IRIF characteristics - permanent cell lines (NHDFs, U87) and primary cell cultures prepared from tumors and adjacent normal tissues of head and neck cancer patients. The cells were dosed with 0.5-8 Gy-gamma-rays and fixed at multiple (0-24 h) postirradiation times. Under all circumstances, DeepFoci quantified the number of IRIFs with the highest accuracy among current advanced algorithms. Moreover, while the detection error of DeepFoci remained comparable to the variability between two experienced experts, the software maintained its sensitivity and fidelity across dramatically different IRIF counts per nucleus. In addition, information was extracted on IRIF 3D morphometric features and repair protein colocalization within IRIFs. This approach allowed multiparameter IRIF categorization of single- or multichannel data, thereby refining the analysis of DSB repair processes and classification of patient tumors, with the potential to identify specific cell subclones. The developed software improves IRIF quantification for various practical applications (radiotherapy monitoring, biodosimetry, etc.) and opens the door to advanced DSB focus analysis and, in turn, a better understanding of (radiation-induced) DNA damage and repair. (C) 2021 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology.

Návaznosti

LM2018140, projekt VaV

Název: e-Infrastruktura CZ (Akronym: e-INFRA CZ)

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, e-Infrastruktura CZ

MUNI/A/1307/2019, interní kód MU

Název: Kardiovaskulární systém od A do Z (Akronym: KAVASAZ)

Investor: Masarykova univerzita, Kardiovaskulární systém od A do Z, DO R. 2020_Kategorie A - Specifický výzkum - Studentské výzkumné projekty

MUNI/A/1453/2019, interní kód MU

Název: Příspěvek ke studiu molekulární, buněčné, tkáňové a systémové patofyziologie vybraných komplexních nemocí

Investor: Masarykova univerzita, Příspěvek ke studiu molekulární, buněčné, tkáňové a systémové patofyziologie vybraných komplexních nemocí, DO R. 2020_Kategorie A - Specifický výzkum - Studentské výzkumné projekty

ROZV/28/LF26/2020, interní kód MU

Název: Metoda analýzy dvouvláknových zlomů DNA na základě konfokální mikrosi

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, Metoda analýzy dvouvláknových zlomů DNA na základě konfokální mikroskopie, Interní rozvojové projekty

Citovat

VIČAR, Tomáš, Jaromír GUMULEC, Radim KOLAR, Olga KOPECNA, Eva PAGACOVA, Iva FALKOVA a Martin FALK. DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci. Computational and Structural Biotechnology Journal. Amsterdam: Elsevier, 2021, roč. 19, č. 2021, s. 6465-6480. ISSN 2001-0370. Dostupné z: https://dx.doi.org/10.1016/j.csbj.2021.11.019.

@article{1826222,
   author = {Vičar, Tomáš and Gumulec, Jaromír and Kolar, Radim and Kopecna, Olga and Pagacova, Eva and Falkova, Iva and Falk, Martin},
   article_location = {Amsterdam},
   article_number = {2021},
   doi = {http://dx.doi.org/10.1016/j.csbj.2021.11.019},
   keywords = {DNA Damage and Repair; Ionizing Radiation-Induced Foci (IRIFs); Biodosimetry; Deep Learning; Convolutional Neural Network; Morphometry; Confocal Microscopy; Image Analysis},
   language = {eng},
   issn = {2001-0370},
   journal = {Computational and Structural Biotechnology Journal},
   title = {DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci},
   url = {https://www.sciencedirect.com/science/article/pii/S2001037021004840?via%3Dihub},
   volume = {19},
   year = {2021}
}

TY  - JOUR
ID  - 1826222
AU  - Vičar, Tomáš - Gumulec, Jaromír - Kolar, Radim - Kopecna, Olga - Pagacova, Eva - Falkova, Iva - Falk, Martin
PY  - 2021
TI  - DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci
JF  - Computational and Structural Biotechnology Journal
VL  - 19
IS  - 2021
SP  - 6465-6480
EP  - 6465-6480
PB  - Elsevier
SN  - 20010370
KW  - DNA Damage and Repair
KW  - Ionizing Radiation-Induced Foci (IRIFs)
KW  - Biodosimetry
KW  - Deep Learning
KW  - Convolutional Neural Network
KW  - Morphometry
KW  - Confocal Microscopy
KW  - Image Analysis
UR  - https://www.sciencedirect.com/science/article/pii/S2001037021004840?via%3Dihub
N2  - DNA double-strand breaks (DSBs), marked by ionizing radiation-induced (repair) foci (IRIFs), are the most serious DNA lesions and are dangerous to human health. IRIF quantification based on confocal microscopy represents the most sensitive and gold-standard method in radiation biodosimetry and allows research on DSB induction and repair at the molecular and single-cell levels. In this study, we introduce DeepFoci - a deep learning-based fully automatic method for IRIF counting and morphometric analysis. DeepFoci is designed to work with 3D multichannel data (trained for 53BP1 and gamma H2AX) and uses U-Net for nucleus segmentation and IRIF detection, together with maximally stable extremal region-based IRIF segmentation. The proposed method was trained and tested on challenging datasets consisting of mixtures of nonir-radiated and irradiated cells of different types and IRIF characteristics - permanent cell lines (NHDFs, U87) and primary cell cultures prepared from tumors and adjacent normal tissues of head and neck cancer patients. The cells were dosed with 0.5-8 Gy-gamma-rays and fixed at multiple (0-24 h) postirradiation times. Under all circumstances, DeepFoci quantified the number of IRIFs with the highest accuracy among current advanced algorithms. Moreover, while the detection error of DeepFoci remained comparable to the variability between two experienced experts, the software maintained its sensitivity and fidelity across dramatically different IRIF counts per nucleus. In addition, information was extracted on IRIF 3D morphometric features and repair protein colocalization within IRIFs. This approach allowed multiparameter IRIF categorization of single- or multichannel data, thereby refining the analysis of DSB repair processes and classification of patient tumors, with the potential to identify specific cell subclones. The developed software improves IRIF quantification for various practical applications (radiotherapy monitoring, biodosimetry, etc.) and opens the door to advanced DSB focus analysis and, in turn, a better understanding of (radiation-induced) DNA damage and repair. (C) 2021 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology.
ER  -

VIČAR, Tomáš, Jaromír GUMULEC, Radim KOLAR, Olga KOPECNA, Eva PAGACOVA, Iva FALKOVA a Martin FALK. DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci. \textit{Computational and Structural Biotechnology Journal}. Amsterdam: Elsevier, 2021, roč.~19, č.~2021, s.~6465-6480. ISSN~2001-0370. Dostupné z: https://dx.doi.org/10.1016/j.csbj.2021.11.019.

Podrobný výpis o publikaci