scholarly journals YeastMate: Neural network-assisted segmentation of mating and budding events in S. cerevisiae

2021 ◽  
Author(s):  
David Bunk ◽  
Julian Moriasy ◽  
Felix Thoma ◽  
Christopher Jakubke ◽  
Christof Osman ◽  
...  
Keyword(s):  
Neural Network ◽  
Saccharomyces Cerevisiae ◽  
Deep Learning ◽  
Source Code ◽  
Automated Detection ◽  
Daughter Cells ◽  
Mother And Daughter ◽  
User Friendly ◽  
Microscopy Images

Here, we introduce YeastMate, a user-friendly deep learning- based application for automated detection and segmentation of Saccharomyces cerevisiae cells and their mating and budding events in microscopy images. We build upon Mask R-CNN with a custom segmentation head for the subclassification of mother and daughter cells during lifecycle transitions. YeastMate can be used directly as a Python library or through a stand-alone GUI application and a Fiji plugin as easy to use frontends. The source code for YeastMate is freely available at https://github.com/hoerlteam/YeastMate under the MIT license. We offer packaged installers for our whole software stack for Windows, macOS and Linux. A detailed user guide is available at https://yeastmate.readthedocs.io.

scholarly journals Orchard Mapping with Deep Learning Semantic Segmentation

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3813
Author(s):  
Athanasios Anagnostis ◽  
Aristotelis C. Tagarakis ◽  
Dimitrios Kateris ◽  
Vasileios Moysiadis ◽  
Claus Grøn Sørensen ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Semantic Segmentation ◽  
Automated Detection ◽  
Aerial Images ◽  
Training Dataset ◽  
Field Boundary ◽  
Different Seasons ◽  
Detection And Localization ◽  
Different Levels

This study aimed to propose an approach for orchard trees segmentation using aerial images based on a deep learning convolutional neural network variant, namely the U-net network. The purpose was the automated detection and localization of the canopy of orchard trees under various conditions (i.e., different seasons, different tree ages, different levels of weed coverage). The implemented dataset was composed of images from three different walnut orchards. The achieved variability of the dataset resulted in obtaining images that fell under seven different use cases. The best-trained model achieved 91%, 90%, and 87% accuracy for training, validation, and testing, respectively. The trained model was also tested on never-before-seen orthomosaic images or orchards based on two methods (oversampling and undersampling) in order to tackle issues with out-of-the-field boundary transparent pixels from the image. Even though the training dataset did not contain orthomosaic images, it achieved performance levels that reached up to 99%, demonstrating the robustness of the proposed approach.

scholarly journals Encoding Health Records into Pathway Representations for Deep Learning

2021 ◽  
Author(s):  
Marco Luca Sbodio ◽  
Natasha Mulligan ◽  
Stefanie Speichert ◽  
Vanessa Lopez ◽  
Joao Bettencourt-Silva
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Source Code ◽  
Training Dataset ◽  
Health Records ◽  
Learning Tasks ◽  
Patient Pathways ◽  
Computational Resources ◽  
The Impact

There is a growing trend in building deep learning patient representations from health records to obtain a comprehensive view of a patient’s data for machine learning tasks. This paper proposes a reproducible approach to generate patient pathways from health records and to transform them into a machine-processable image-like structure useful for deep learning tasks. Based on this approach, we generated over a million pathways from FAIR synthetic health records and used them to train a convolutional neural network. Our initial experiments show the accuracy of the CNN on a prediction task is comparable or better than other autoencoders trained on the same data, while requiring significantly less computational resources for training. We also assess the impact of the size of the training dataset on autoencoders performances. The source code for generating pathways from health records is provided as open source.

scholarly journals Deep Transfer Learning for Source Code Modeling

2020 ◽  
Vol 30 (05) ◽  
pp. 649-668
Author(s):  
Yasir Hussain ◽  
Zhiqiu Huang ◽  
Yu Zhou ◽  
Senzhang Wang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Transfer Learning ◽  
Related Problem ◽  
State Of The Art ◽  
Source Code ◽  
Traditional Learning ◽  
Learning Models ◽  
Modeling And Analysis ◽  
F Measure

In recent years, deep learning models have shown great potential in source code modeling and analysis. Generally, deep learning-based approaches are problem-specific and data-hungry. A challenging issue of these approaches is that they require training from scratch for a different related problem. In this work, we propose a transfer learning-based approach that significantly improves the performance of deep learning-based source code models. In contrast to traditional learning paradigms, transfer learning can transfer the knowledge learned in solving one problem into another related problem. First, we present two recurrent neural network-based models RNN and GRU for the purpose of transfer learning in the domain of source code modeling. Next, via transfer learning, these pre-trained (RNN and GRU) models are used as feature extractors. Then, these extracted features are combined into attention learner for different downstream tasks. The attention learner leverages from the learned knowledge of pre-trained models and fine-tunes them for a specific downstream task. We evaluate the performance of the proposed approach with extensive experiments with the source code suggestion task. The results indicate that the proposed approach outperforms the state-of-the-art models in terms of accuracy, precision, recall and F-measure without training the models from scratch.

scholarly journals Cell cycle phases in the unequal mother/daughter cell cycles of Saccharomyces cerevisiae.

1984 ◽  
Vol 4 (11) ◽  
pp. 2529-2531 ◽  
Author(s):  
B J Brewer ◽  
E Chlebowicz-Sledziewska ◽  
W L Fangman
Keyword(s):  
Saccharomyces Cerevisiae ◽  
Cell Cycle ◽  
Cycle Phase ◽  
Cell Cycle Time ◽  
Yeast Saccharomyces Cerevisiae ◽  
Cell Cycles ◽  
Daughter Cells ◽  
Mother And Daughter ◽  
Cell Cycle Phases ◽  
Mother Cells

During cell division in the yeast Saccharomyces cerevisiae mother cells produce buds (daughter cells) which are smaller and have longer cell cycles. We performed experiments to compare the lengths of cell cycle phases in mothers and daughters. As anticipated from earlier indirect observations, the longer cell cycle time of daughter cells is accounted for by a longer G1 interval. The S-phase and the G2-phase are of the same duration in mother and daughter cells. An analysis of five isogenic strains shows that cell cycle phase lengths are independent of cell ploidy and mating type.

scholarly journals Roles of the 2 microns gene products in stable maintenance of the 2 microns plasmid of Saccharomyces cerevisiae.

1987 ◽  
Vol 7 (10) ◽  
pp. 3566-3573 ◽  
Author(s):  
A E Reynolds ◽  
A W Murray ◽  
J W Szostak
Keyword(s):  
Saccharomyces Cerevisiae ◽  
Copy Number ◽  
Gene Products ◽  
Site Specific ◽  
Cell Cycles ◽  
Daughter Cells ◽  
Repeat Sequences ◽  
Copy Numbers ◽  
Mother And Daughter ◽  
Stable Maintenance

We have examined the replication and segregation of the Saccharomyces cerevisiae 2 microns circle. The amplification of the plasmid at low copy numbers requires site-specific recombination between the 2 microns inverted repeat sequences catalyzed by the plasmid-encoded FLP gene. No other 2 microns gene products are required. The overexpression of FLP in a strain carrying endogenous 2 microns leads to uncontrolled plasmid replication, longer cell cycles, and cell death. Two different assays show that the level of Flp activity decreases with increasing 2 microns copy number. This regulation requires the products of the REP1 and REP2 genes. These gene products also act together to ensure that 2 microns molecules are randomly segregated between mother and daughter cells at cell division.

scholarly journals The fitness cost of mismatch repair mutators in Saccharomyces cerevisiae: partitioning the mutational load

2019 ◽  
Author(s):  
Benjamin Galeota-Sprung ◽  
Breanna Guindon ◽  
Paul Sniegowski
Keyword(s):  
Saccharomyces Cerevisiae ◽  
Mismatch Repair ◽  
Separate Experiment ◽  
Mutational Load ◽  
Deleterious Mutations ◽  
Competitive Fitness ◽  
Daughter Cells ◽  
Point Estimates ◽  
Mother And Daughter ◽  
Mean Fitness

AbstractMutational load is the depression in a population’s mean fitness that results from the continual influx of deleterious mutations. Here, we directly estimate the mutational load in a population of haploid Saccharomyces cerevisiae that are deficient for mismatch repair. We partition the load in haploids into two components. To estimate the load due to nonlethal mutations, we measure the competitive fitness of hundreds of randomly selected clones from both mismatch repair-deficient and - proficient populations. Computation of the mean clone fitness for the mismatch repair-deficient strain permits an estimation of the nonlethal load, and the histogram of fitness provides an interesting visualization of a loaded population. In a separate experiment, in order to estimate the load due to lethal mutations (i.e. the lethal mutation rate), we manipulate thousands of individual pairs of mother and daughter cells and track their fates. These two approaches yield point estimates for the two contributors to load, and the addition of these estimates is nearly equal to the separately measured short-term competitive fitness deficit for the mismatch repair-deficient strain. This correspondence suggests that there is no need to invoke direct fitness effects to explain the fitness difference between mismatch repair-deficient and - proficient strains. Assays in diploids are consistent with deleterious mutations in diploids tending towards recessivity. These results enhance our understanding of mutational load, a central population genetics concept, and we discuss their implications for the evolution of mutation rates.

scholarly journals SPOROGENIC ACTIVITIES OF MOTHER AND DAUGHTER CELLS IN SACCHAROMYCES CEREVISIAE

1970 ◽  
Vol 16 (4) ◽  
pp. 347-350 ◽  
Author(s):  
TOMOMICHI YANAGITA ◽  
MORIMASA YAGISAWA ◽  
SHINSHI OISHI ◽  
NOBUNDO SANDO ◽  
TSUNEJI SUTO
Keyword(s):  
Saccharomyces Cerevisiae ◽  
Daughter Cells ◽  
Mother And Daughter

scholarly journals Artifically intelligent physics solver: This AI understands Newtons law

2019 ◽  
Vol 7 (1) ◽  
pp. 22-28
Author(s):  
V. Nirmala ◽  
◽  
A. Rajagopal ◽  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Neural Network Model ◽  
Language Model ◽  
Source Code ◽  
Learning Module ◽  
Learning Technique ◽  
Third Law ◽  
The Given ◽  
Physics Problems

implemented a working prototype of a Deep Learning module that seem to understand Newton’s third law of motion. The networks. In this paper, a Google BERT neural network model was trained using transfer learning technique on a synthetic dataset of simple physics problems within the scope of solving Newton’s third law problems that requires understanding of concepts such as action and reaction, magnitude and direction forces, simple concepts of vectors in physics problems. The of Netwon’s third law assuming certain boundaries on the language model of the word problems. A working prototype of this AI can be accessed at the given website. This paper also contributes the source code for reproducible results. This novel idea can be extended to more science topics. Applications of this interdisciplinary area of AI and physics have impact not just in areas of robotics and computational physics, but also in how science uses AI in the future. In future, more areas of .

Sign in / Sign up

Export Citation Format

Share Document