scholarly journals VIDHOP, viral host prediction with deep learning

2020 ◽  
Author(s):  
Florian Mock ◽  
Adrian Viehweger ◽  
Emanuel Barth ◽  
Manja Marz
Keyword(s):  
Deep Learning ◽  
Viral Genome ◽  
Influenza A ◽  
Deep Neural Networks ◽  
Supplementary Information ◽  
Learning Approach ◽  
Virus Species ◽  
Genome Sequences ◽  
Rotavirus A ◽  
The Core

Abstract Motivation Zoonosis, the natural transmission of infections from animals to humans, is a far-reaching global problem. The recent outbreaks of Zikavirus, Ebolavirus and Coronavirus are examples of viral zoonosis, which occur more frequently due to globalization. In case of a virus outbreak, it is helpful to know which host organism was the original carrier of the virus to prevent further spreading of viral infection. Recent approaches aim to predict a viral host based on the viral genome, often in combination with the potential host genome and arbitrarily selected features. These methods are limited in the number of different hosts they can predict or the accuracy of the prediction. Results Here, we present a fast and accurate deep learning approach for viral host prediction, which is based on the viral genome sequence only. We tested our deep neural network (DNN) on three different virus species (influenza A virus, rabies lyssavirus and rotavirus A). We achieved for each virus species an AUC between 0.93 and 0.98, allowing highly accurate predictions while using only fractions (100–400 bp) of the viral genome sequences. We show that deep neural networks are suitable to predict the host of a virus, even with a limited amount of sequences and highly unbalanced available data. The trained DNNs are the core of our virus–host prediction tool VIrus Deep learning HOst Prediction (VIDHOP). VIDHOP also allows the user to train and use models for other viruses. Availability and implementation VIDHOP is freely available under https://github.com/flomock/vidhop. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals VIDHOP, viral host prediction with Deep Learning

2019 ◽  
Author(s):  
Florian Mock ◽  
Adrian Viehweger ◽  
Emanuel Barth ◽  
Manja Marz
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Viral Genome ◽  
Influenza A ◽  
Deep Neural Networks ◽  
Ebola Virus ◽  
Training Data ◽  
Supplementary Information ◽  
Virus Species ◽  
Rotavirus A

AbstractMotivationZoonosis, the natural transmission of infections from animals to humans, is a far-reaching global problem. The recent outbreaks of Zika virus, Ebola virus and Corona virus are examples of viral zoonosis, which occur more frequently due to globalization. In the case of a virus outbreak, it is helpful to know which host organism was the original carrier of the virus. Once the reservoir or intermediate host is known, it can be isolated to prevent further spreading of the viral infection. Recent approaches aim to predict a viral host based on the viral genome, often in combination with the potential host genome and arbitrarily selected features. These methods have a clear limitation in either the number of different hosts they can predict or the accuracy of their prediction.ResultsHere, we present a fast and accurate deep learning approach for viral host prediction, which is based on the viral genome sequence only. To ensure a high prediction accuracy, we developed an effective selection approach for the training data to avoid biases due to a highly unbalanced number of known sequences per virus-host combinations. We tested our deep neural network on three different virus species (influenza A, rabies lyssavirus, rotavirus A). We reached for each virus species an AUG between 0.93 and 0.98, outperforming previous approaches and allowing highly accurate predictions while only using fractions (100-400 bp) of the viral genome sequences. We show that deep neural networks are suitable to predict the host of a virus, even with a limited amount of sequences and highly unbalanced available data. The deep neural networks trained for this approach build the core of the virus-host predicting tool VIDHOP (Virus Deep learning HOst Prediction).AvailabilityThe trained models for the prediction of the host for the viruses influenza A, rabies lyssavirus, rotavirus A are implemented in the tool VIDHOP. This tool is freely available under https://github.com/flomock/vidhop.Supplementary informationSupplementary data are available at DOI 10.17605/OSF.IO/UXT7N

scholarly journals Representing Deep Neural Networks Latent Space Geometries with Graphs

Algorithms ◽  
2021 ◽  
Vol 14 (2) ◽  
pp. 39
Author(s):  
Carlos Lassance ◽  
Vincent Gripon ◽  
Antonio Ortega
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Learning ◽  
Objective Function ◽  
Learning Process ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
The Core ◽  
Learning Tasks ◽  
Latent Space

Deep Learning (DL) has attracted a lot of attention for its ability to reach state-of-the-art performance in many machine learning tasks. The core principle of DL methods consists of training composite architectures in an end-to-end fashion, where inputs are associated with outputs trained to optimize an objective function. Because of their compositional nature, DL architectures naturally exhibit several intermediate representations of the inputs, which belong to so-called latent spaces. When treated individually, these intermediate representations are most of the time unconstrained during the learning process, as it is unclear which properties should be favored. However, when processing a batch of inputs concurrently, the corresponding set of intermediate representations exhibit relations (what we call a geometry) on which desired properties can be sought. In this work, we show that it is possible to introduce constraints on these latent geometries to address various problems. In more detail, we propose to represent geometries by constructing similarity graphs from the intermediate representations obtained when processing a batch of inputs. By constraining these Latent Geometry Graphs (LGGs), we address the three following problems: (i) reproducing the behavior of a teacher architecture is achieved by mimicking its geometry, (ii) designing efficient embeddings for classification is achieved by targeting specific geometries, and (iii) robustness to deviations on inputs is achieved via enforcing smooth variation of geometry between consecutive latent spaces. Using standard vision benchmarks, we demonstrate the ability of the proposed geometry-based methods in solving the considered problems.

scholarly journals DeepFlu: a deep learning approach for forecasting symptomatic influenza A infection based on pre-exposure gene expression

2021 ◽  
pp. 106495
Author(s):  
Anna Zan ◽  
Zhong-Ru Xie ◽  
Yi-Chen Hsu ◽  
Yu-Hao Chen ◽  
Tsung-Hsien Lin ◽  
...  
Keyword(s):  
Gene Expression ◽  
Deep Learning ◽  
Influenza A ◽  
Learning Approach

scholarly journals Beyond Core Object Recognition: Recurrent processes account for object recognition under occlusion

2018 ◽  
Author(s):  
Karim Rajaei ◽  
Yalda Mohsenzadeh ◽  
Reza Ebrahimpour ◽  
Seyed-Mahdi Khaligh-Razavi
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Object Recognition ◽  
Human Brain ◽  
Deep Neural Networks ◽  
Temporal Dynamics ◽  
Computational Modelling ◽  
Mechanistic Explanation ◽  
The Core ◽  
Level Performance

AbstractCore object recognition, the ability to rapidly recognize objects despite variations in their appearance, is largely solved through the feedforward processing of visual information. Deep neural networks are shown to achieve human-level performance in these tasks, and explain the primate brain representation. On the other hand, object recognition under more challenging conditions (i.e. beyond the core recognition problem) is less characterized. One such example is object recognition under occlusion. It is unclear to what extent feedforward and recurrent processes contribute in object recognition under occlusion. Furthermore, we do not know whether the conventional deep neural networks, such as AlexNet, which were shown to be successful in solving core object recognition, can perform similarly well in problems that go beyond the core recognition. Here, we characterize neural dynamics of object recognition under occlusion, using magnetoencephalography (MEG), while participants were presented with images of objects with various levels of occlusion. We provide evidence from multivariate analysis of MEG data, behavioral data, and computational modelling, demonstrating an essential role for recurrent processes in object recognition under occlusion. Furthermore, the computational model with local recurrent connections, used here, suggests a mechanistic explanation of how the human brain might be solving this problem.Author SummaryIn recent years, deep-learning-based computer vision algorithms have been able to achieve human-level performance in several object recognition tasks. This has also contributed in our understanding of how our brain may be solving these recognition tasks. However, object recognition under more challenging conditions, such as occlusion, is less characterized. Temporal dynamics of object recognition under occlusion is largely unknown in the human brain. Furthermore, we do not know if the previously successful deep-learning algorithms can similarly achieve human-level performance in these more challenging object recognition tasks. By linking brain data with behavior, and computational modeling, we characterized temporal dynamics of object recognition under occlusion, and proposed a computational mechanism that explains both behavioral and the neural data in humans. This provides a plausible mechanistic explanation for how our brain might be solving object recognition under more challenging conditions.

scholarly journals Assessment of bilateral knee pain from MR imaging using deep neural networks

2018 ◽  
Author(s):  
Gary H. Chang ◽  
David T. Felson ◽  
Shangran Qiu ◽  
Terence D. Capellini ◽  
Vijaya B. Kolachalama
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Neural Networks ◽  
Deep Learning ◽  
Knee Pain ◽  
Deep Neural Networks ◽  
Learning Approach ◽  
Mr Images ◽  
Bilateral Knee ◽  
Machine Learning Approach

ABSTRACTBackground and objectiveIt remains difficult to characterize pain in knee joints with osteoarthritis solely by radiographic findings. We sought to understand how advanced machine learning methods such as deep neural networks can be used to analyze raw MRI scans and predict bilateral knee pain, independent of other risk factors.MethodsWe developed a deep learning framework to associate information from MRI slices taken from the left and right knees of subjects from the Osteoarthritis Initiative with bilateral knee pain. Model training was performed by first extracting features from two-dimensional (2D) sagittal intermediate-weighted turbo spin echo slices. The extracted features from all the 2D slices were subsequently combined to directly associate using a fused deep neural network with the output of interest as a binary classification problem.ResultsThe deep learning model resulted in predicting bilateral knee pain on test data with 70.1% mean accuracy, 51.3% mean sensitivity, and 81.6% mean specificity. Systematic analysis of the predictions on the test data revealed that the model performance was consistent across subjects of different Kellgren-Lawrence grades.ConclusionThe study demonstrates a proof of principle that a machine learning approach can be applied to associate MR images with bilateral knee pain.SIGNIFICANCE AND INNOVATIONKnee pain is typically considered as an early indicator of osteoarthritis (OA) risk. Emerging evidence suggests that MRI changes are linked to pre-clinical OA, thus underscoring the need for building image-based models to predict knee pain. We leveraged a state-of-the-art machine learning approach to associate raw MR images with bilateral knee pain, independent of other risk factors.

Subtype-GAN: a deep learning approach for integrative cancer subtyping of multi-omics data

2021 ◽  
Author(s):  
Hai Yang ◽  
Rui Chen ◽  
Dongdong Li ◽  
Zhe Wang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Latent Variables ◽  
Supplementary Information ◽  
Learning Approach ◽  
Data Sets ◽  
Omics Data ◽  
Data Set ◽  
Benchmark Data ◽  
Cancer Pathogenesis

Abstract Motivation The discovery of cancer subtyping can help explore cancer pathogenesis, determine clinical actionability in treatment, and improve patients' survival rates. However, due to the diversity and complexity of multi-omics data, it is still challenging to develop integrated clustering algorithms for tumor molecular subtyping. Results We propose Subtype-GAN, a deep adversarial learning approach based on the multiple-input multiple-output neural network to model the complex omics data accurately. With the latent variables extracted from the neural network, Subtype-GAN uses consensus clustering and the Gaussian Mixture model to identify tumor samples' molecular subtypes. Compared with other state-of-the-art subtyping approaches, Subtype-GAN achieved outstanding performance on the benchmark data sets consisting of ∼4,000 TCGA tumors from 10 types of cancer. We found that on the comparison data set, the clustering scheme of Subtype-GAN is not always similar to that of the deep learning method AE but is identical to that of NEMO, MCCA, VAE, and other excellent approaches. Finally, we applied Subtype-GAN to the BRCA data set and automatically obtained the number of subtypes and the subtype labels of 1031 BRCA tumors. Through the detailed analysis, we found that the identified subtypes are clinically meaningful and show distinct patterns in the feature space, demonstrating the practicality of Subtype-GAN. Availability The source codes, the clustering results of Subtype-GAN across the benchmark data sets are available at https://github.com/haiyang1986/Subtype-GAN. Supplementary information Supplementary data are available at Bioinformatics online.

CoCoNet: an efficient deep learning tool for viral metagenome binning

2021 ◽  
Author(s):  
Cédric G Arisdakessian ◽  
Olivia D Nigro ◽  
Grieg F Steward ◽  
Guylaine Poisson ◽  
Mahdi Belcaid
Keyword(s):  
Deep Learning ◽  
Viral Genome ◽  
High Performance ◽  
Source Code ◽  
Supplementary Information ◽  
Biological Processes ◽  
Bacterial Genomes ◽  
Large Dataset ◽  
Sequence Composition ◽  
Rigorous Framework

Abstract Motivation Metagenomic approaches hold the potential to characterize microbial communities and unravel the intricate link between the microbiome and biological processes. Assembly is one of the most critical steps in metagenomics experiments. It consists of transforming overlapping DNA sequencing reads into sufficiently accurate representations of the community’s genomes. This process is computationally difficult and commonly results in genomes fragmented across many contigs. Computational binning methods are used to mitigate fragmentation by partitioning contigs based on their sequence composition, abundance or chromosome organization into bins representing the community’s genomes. Existing binning methods have been principally tuned for bacterial genomes and do not perform favorably on viral metagenomes. Results We propose Composition and Coverage Network (CoCoNet), a new binning method for viral metagenomes that leverages the flexibility and the effectiveness of deep learning to model the co-occurrence of contigs belonging to the same viral genome and provide a rigorous framework for binning viral contigs. Our results show that CoCoNet substantially outperforms existing binning methods on viral datasets. Availability and implementation CoCoNet was implemented in Python and is available for download on PyPi (https://pypi.org/). The source code is hosted on GitHub at https://github.com/Puumanamana/CoCoNet and the documentation is available at https://coconet.readthedocs.io/en/latest/index.html. CoCoNet does not require extensive resources to run. For example, binning 100k contigs took about 4 h on 10 Intel CPU Cores (2.4 GHz), with a memory peak at 27 GB (see Supplementary Fig. S9). To process a large dataset, CoCoNet may need to be run on a high RAM capacity server. Such servers are typically available in high-performance or cloud computing settings. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Combining content and social features in a deep learning approach to Vietnamese email prioritization

2021 ◽  
Author(s):  
Ha Thanh Nguyen ◽  
Quan Dinh Dang ◽  
Anh Quang Tran
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
New Method ◽  
Learning Approach ◽  
Field Of Study ◽  
Learning Techniques ◽  
Email Overload ◽  
Social Features

The email overload problem has been discussed in numerous email-related studies. One of the possible solutions to this problem is email prioritization, which is the act of automatically predicting the importance levels of received emails and sorting the user’s inbox accordingly. Several learning-based methods have been proposed to address the email prioritization problem using content features as well as social features. Although these methods have laid the foundation works in this field of study, the reported performance is far from being practical. Recent works on deep neural networks have achieved good results in various tasks. In this paper, the authors propose a novel email prioritization model which incorporates several deep learning techniques and uses a combination of both content features and social features from email data. This method targets Vietnamese emails and is tested against a self-built Vietnamese email corpus. Conducted experiments explored the effects of different model configurations and compared the effectiveness of the new method to that of a previous work.

scholarly journals A deep learning approach to pattern recognition for short DNA sequences

2018 ◽  
Author(s):  
Akosua Busia ◽  
George E. Dahl ◽  
Clara Fannjiang ◽  
David H. Alexander ◽  
Elizabeth Dorfman ◽  
...  
Keyword(s):  
Deep Learning ◽  
Dna Sequences ◽  
Distinct Species ◽  
Training Data ◽  
Supplementary Information ◽  
Learning Approach ◽  
Biological Sequences ◽  
Public Research ◽  
Species Classification ◽  
Link Type

AbstractMotivationInferring properties of biological sequences--such as determining the species-of-origin of a DNA sequence or the function of an amino-acid sequence--is a core task in many bioinformatics applications. These tasks are often solved using string-matching to map query sequences to labeled database sequences or via Hidden Markov Model-like pattern matching. In the current work we describe and assess an deep learning approach which trains a deep neural network (DNN) to predict database-derived labels directly from query sequences.ResultsWe demonstrate this DNN performs at state-of-the-art or above levels on a difficult, practically important problem: predicting species-of-origin from short reads of 16S ribosomal DNA. When trained on 16S sequences of over 13,000 distinct species, our DNN achieves read-level species classification accuracy within 2.0% of perfect memorization of training data, and produces more accurate genus-level assignments for reads from held-out species thank-mer, alignment, and taxonomic binning baselines. Moreover, our models exhibit greater robustness than these existing approaches to increasing noise in the query sequences. Finally, we show that these DNNs perform well on experimental 16S mock community dataset. Overall, our results constitute a first step towards our long-term goal of developing a general-purpose deep learning approach to predicting meaningful labels from short biological sequences.AvailabilityTensorFlow training code is available through GitHub (https://github.com/tensorflow/models/tree/master/research). Data in TensorFlow TFRecord format is available on Google Cloud Storage (gs://brain-genomics-public/research/seq2species/)[email protected] informationSupplementary data are available in a separate document.

scholarly journals Using deep neural networks to detect complex spikes of cerebellar Purkinje cells

2020 ◽  
Vol 123 (6) ◽  
pp. 2217-2234
Author(s):  
Akshay Markanday ◽  
Joachim Bellet ◽  
Marie E. Bellet ◽  
Junya Inoue ◽  
Ziad M. Hafed ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Purkinje Cell ◽  
Purkinje Cells ◽  
Deep Neural Networks ◽  
Learning Approach ◽  
Sorting Algorithms ◽  
Accuracy Level ◽  
Cerebellar Purkinje Cells ◽  
Complex Spikes

Purkinje cell “complex spikes,” fired at perplexingly low rates, play a crucial role in cerebellum-based motor learning. Careful interpretations of these spikes require manually detecting them, since conventional online or offline spike sorting algorithms are optimized for classifying much simpler waveform morphologies. We present a novel deep learning approach for identifying complex spikes, which also measures additional relevant neurophysiological features, with an accuracy level matching that of human experts yet with very little time expenditure.

Sign in / Sign up

Export Citation Format

Share Document