Interphase Cell Cycle Staging using Deep Learning *

The cell cycle is an important process in cellular life. In recent years, some image processing methods have been developed to determine the cell cycle stages of individual cells. However, in most of these methods, cells have to be segmented, and their features need to be extracted. During feature extraction, some important information may be lost, resulting in lower classification accuracy. Thus, we used a deep learning method to retain all cell features. In order to solve the problems surrounding insufficient numbers of original images and the imbalanced distribution of original images, we used the Wasserstein generative adversarial network-gradient penalty (WGAN-GP) for data augmentation. At the same time, a residual network (ResNet) was used for image classification. ResNet is one of the most used deep learning classification networks. The classification accuracy of cell cycle images was achieved more effectively with our method, reaching 83.88%. Compared with an accuracy of 79.40% in previous experiments, our accuracy increased by 4.48%. Another dataset was used to verify the effect of our model and, compared with the accuracy from previous results, our accuracy increased by 12.52%. The results showed that our new cell cycle image classification system based on WGAN-GP and ResNet is useful for the classification of imbalanced images. Moreover, our method could potentially solve the low classification accuracy in biomedical images caused by insufficient numbers of original images and the imbalanced distribution of original images.

Download Full-text

Timing of nuclear and kinetoplast DNA replication and early morphological events in the cell cycle of Trypanosoma brucei

Journal of Cell Science ◽

10.1242/jcs.95.1.49 ◽

1990 ◽

Vol 95 (1) ◽

pp. 49-57 ◽

Cited By ~ 1

Author(s):

R. Woodward ◽

K. Gull

Keyword(s):

Cell Cycle ◽

Dna Replication ◽

Trypanosoma Brucei ◽

Basal Body ◽

Unit Cell ◽

Kinetoplast Dna ◽

Interphase Cell ◽

Body Formation ◽

Timing Of Initiation ◽

Single Mitochondrion

We have used immunofluorescent detection of 5-bromo-2-deoxyuridine-substituted DNA in order to determine the timing of initiation and the duration of nuclear and kinetoplast S-phases within the procyclic stage of the Trypanosoma brucei cell cycle. Both nuclear and kinetoplast S-phases were shown to be periodic, occupying 0.18 and 0.12 of the unit cell cycle, respectively. In addition, initiation of both of these S-phases were in approximate synchrony, differing by only 0.03 of the unit cell cycle. We have also used a monoclonal antibody that recognises the basal bodies of T. brucei in order to visualise cells possessing a new pro-basal body and hence determine the time of pro-basal body formation within the cell cycle. Pro-basal body formation occurred within a few minutes of the initiation of nuclear S-phase, at 0.41 of the unit cell cycle. This provides detection of the earliest known cell cycle event in T. brucei at the level of the light microscope. Cell cycle events including initiation of nuclear and kinetoplast DNA replication and pro-basal body formation may be strictly coordinated in T. brucei in order to maintain the precise single-mitochondrion (kinetoplast), singleflagellum status of the interphase cell.

Download Full-text

Interphase Cell Cycle Dynamics of a Late-Replicating, Heterochromatic Homogeneously Staining Region: Precise Choreography of Condensation/Decondensation and Nuclear Positioning

The Journal of Cell Biology ◽

10.1083/jcb.140.5.975 ◽

1998 ◽

Vol 140 (5) ◽

pp. 975-989 ◽

Cited By ~ 123

Author(s):

Gang Li ◽

Gail Sudlow ◽

Andrew S. Belmont

Keyword(s):

Cell Cycle ◽

Dna Replication ◽

Dna Sequences ◽

Large Scale ◽

S Phase ◽

Interphase Cell ◽

Compact Mass ◽

Lac Operator ◽

Cell Cycle Dynamics

Recently we described a new method for in situ localization of specific DNA sequences, based on lac operator/repressor recognition (Robinett, C.C., A. Straight, G. Li, C. Willhelm, G. Sudlow, A. Murray, and A.S. Belmont. 1996. J. Cell Biol. 135:1685–1700). We have applied this methodology to visualize the cell cycle dynamics of an ∼90 Mbp, late-replicating, heterochromatic homogeneously staining region (HSR) in CHO cells, combining immunostaining with direct in vivo observations. Between anaphase and early G1, the HSR extends approximately twofold to a linear, ∼0.3-μm-diam chromatid, and then recondenses to a compact mass adjacent to the nuclear envelope. No further changes in HSR conformation or position are seen through mid-S phase. However, HSR DNA replication is preceded by a decondensation and movement of the HSR into the nuclear interior 4–6 h into S phase. During DNA replication the HSR resolves into linear chromatids and then recondenses into a compact mass; this is followed by a third extension of the HSR during G2/ prophase. Surprisingly, compaction of the HSR is extremely high at all stages of interphase. Preliminary ultrastructural analysis of the HSR suggests at least three levels of large-scale chromatin organization above the 30-nm fiber.

Download Full-text

Cell cycle regulation of the p34cdc2/p33cdk2-activating kinase p40MO15

Journal of Cell Science ◽

10.1242/jcs.107.10.2789 ◽

1994 ◽

Vol 107 (10) ◽

pp. 2789-2799 ◽

Cited By ~ 7

Author(s):

R.Y. Poon ◽

K. Yamashita ◽

M. Howell ◽

M.A. Ershler ◽

A. Belyavsky ◽

...

Keyword(s):

Cell Cycle ◽

Phosphorylation Site ◽

Mitotic Cell ◽

Phase Cell ◽

Serum Starvation ◽

Interphase Cell ◽

Cell Extracts ◽

M Phase ◽

Equivalent Site ◽

Cycle Regulation

A key component of Cdc2/Cdk2-activating kinase (CAK) is p40MO15, a protein kinase subunit that phosphorylates the T161/T160 residues of p34cdc2/p33cdk2. The level and activity of p40MO15 were essentially constant during cleavage of fertilised Xenopus eggs and in growing mouse 3T3 cells, but serum starvation of these cells reduced both the level and activity of p40MO15. Although the level and activity of endogenous p40MO15 did not vary in the cell cycle, we found that bacterially expressed p40MO15 was activated more rapidly by M-phase cell extracts than by interphase cell extracts. Bacterially expressed p40MO15 was phosphorylated mainly on serine 170 (a p34cdc2 phosphorylation site) by mitotic cell extracts, but mutation of S170 to alanine did not affect the activation of p40MO15, whereas mutation of T176 (the equivalent site to T161/T160 in p34cdc2/p33cdk2) abolished the activation of P40MO15. These studies suggest that the level and activity of p40MO15 is probably not a major determinant of p34cdc2/p33cdk2 activity in the cell cycle, and that the activation of p40MO15 may require phosphorylation on T176.

Download Full-text

Volumetric Segmentation of Cell Cycle Markers in Confocal Images Using Machine Learning and Deep Learning

Frontiers in Plant Science ◽

10.3389/fpls.2020.01275 ◽

2020 ◽

Vol 11 ◽

Author(s):

Faraz Ahmad Khan ◽

Ute Voß ◽

Michael P. Pound ◽

Andrew P. French

Keyword(s):

Machine Learning ◽

Cell Cycle ◽

Deep Learning ◽

Confocal Images ◽

Volumetric Segmentation

Download Full-text

Mutations in alpha- and beta-tubulin affect spindle formation in Chinese hamster ovary cells.

The Journal of Cell Biology ◽

10.1083/jcb.97.4.1055 ◽

1983 ◽

Vol 97 (4) ◽

pp. 1055-1061 ◽

Cited By ~ 21

Author(s):

I Abraham ◽

M Marcus ◽

F Cabral ◽

M M Gottesman

Keyword(s):

Cell Cycle ◽

Cell Lines ◽

Chinese Hamster Ovary ◽

Chinese Hamster ◽

Ovary Cell ◽

Nonpermissive Temperature ◽

Spindle Formation ◽

Interphase Cell ◽

Alpha Tubulin ◽

Mutant Cells

Two Chinese hamster ovary cell lines with mutated beta-tubulins (Grs-2 and Cmd-4) and one that has a mutation in alpha-tubulin (Tax-1) are temperature sensitive for growth at 40.5 degrees C. To determine the functional defect in these mutant cells at the nonpermissive temperature, they were characterized with respect to cell cycle parameters and microtubule organization and function after relatively short periods at 40.5 degrees C. At the nonpermissive temperature all the mutants had normal appearing cytoplasmic microtubules. Premature chromosome condensation analysis failed to show any discrete step in the interphase cell cycle in which these mutants are arrested. These cells, however, show several defects at the nonpermissive temperature that appear related to the function of microtubules during mitosis. Time-lapse studies showed that mitosis was lengthened in the three mutant lines at 40.5 degrees C as compared with the wild-type cells at this temperature, resulting in a higher proportion of cells in mitosis after temperature shift. There was also a large increase in multinucleated cells in mutant populations after incubation at the nonpermissive temperature. Immunofluorescent studies using a monoclonal anti--alpha-tubulin antibody showed that the mutant cells had a high proportion of abnormal spindles at the nonpermissive temperature. The two altered beta-tubulins and the altered alpha-tubulin all were found to cause a similar phenotype at the high temperature that results in mitotic delay, defective cytokinesis, multinucleation, and ultimately, cell death. We conclude that spindle formation is the limiting microtubule function in these mutant cell lines at the nonpermissive temperature and that these cell lines will be of value for the study of the precise role of tubulin in mammalian spindle formation.

Download Full-text

Reconstructing cell cycle and disease progression using deep learning

10.1101/081364 ◽

2016 ◽

Cited By ~ 4

Author(s):

Philipp Eulenberg ◽

Niklas Köhler ◽

Thomas Blasi ◽

Andrew Filby ◽

Anne E. Carpenter ◽

...

Keyword(s):

Cell Cycle ◽

Deep Learning ◽

Disease Progression ◽

Image Data ◽

Image Features ◽

Jurkat Cells ◽

Biological Processes ◽

Deep Convolutional Neural Networks ◽

Recent Approach ◽

Fold Reduction

AbstractWe show that deep convolutional neural networks combined with non-linear dimension reduction enable reconstructing biological processes based on raw image data. We demonstrate this by recon-structing the cell cycle of Jurkat cells and disease progression in diabetic retinopathy. In further analysis of Jurkat cells, we detect and separate a subpopulation of dead cells in an unsupervised manner and, in classifying discrete cell cycle stages, we reach a 6-fold reduction in error rate compared to a recent approach based on boosting on image features. In contrast to previous methods, deep learning based predictions are fast enough for on-the-fly analysis in an imaging flow cytometer.

Download Full-text

Robust classification of cell cycle phase and biological feature extraction by image-based deep learning

Molecular Biology of the Cell ◽

10.1091/mbc.e20-03-0187 ◽

2020 ◽

Vol 31 (13) ◽

pp. 1346-1354 ◽

Cited By ~ 3

Author(s):

Yukiko Nagao ◽

Mika Sakamoto ◽

Takumi Chinen ◽

Yasushi Okada ◽

Daisuke Takao

Keyword(s):

Neural Network ◽

Cell Cycle ◽

Feature Extraction ◽

Deep Learning ◽

Cell Cycle Phase ◽

Data Driven ◽

Cycle Phase ◽

Biological Feature ◽

Biological Features

By applying convolutional neural network-based classifiers, we demonstrate that cell images can be robustly classified according to cell cycle phases. Combined with Grad-CAM analysis, our approach enables us to extract biological features underlying cellular phenomena of interest in an unbiased and data-driven manner.

Download Full-text