scholarly journals Precursor microRNA Identification Using Deep Convolutional Neural Networks

2018 ◽  
Author(s):  
Binh Thanh Do ◽  
Vladimir Golkov ◽  
Göktuğ Erce Gürel ◽  
Daniel Cremers
Keyword(s):  
State Of The Art ◽  
Matrix Representation ◽  
Regulation Of Gene Expression ◽  
Deep Convolutional Neural Networks ◽  
Convolutional Network ◽  
Mirna Identification ◽  
The Neural Network ◽  
Non Coding Rna ◽  
Benchmark Datasets ◽  
Post Transcriptional Regulation

AbstractPrecursor microRNA (pre-miRNA) identification is the basis for identifying microRNAs (miRNAs), which have important roles in post-transcriptional regulation of gene expression. In this paper, we propose a deep learning method to identify whether a small non-coding RNA sequence is a pre-miRNA or not. We outperform state-of-the-art methods on three benchmark datasets, namely the human, cross-species, and new datasets. The key of our method is to use a matrix representation of predicted secondary structure as input to a 2D convolutional network. The neural network extracts optimized features automatically instead of using a large number of handcrafted features as most existing methods do. Code and results are available at https://github.com/peace195/miRNA-identification-conv2D.

scholarly journals MDAN-UNet: Multi-Scale and Dual Attention Enhanced Nested U-Net Architecture for Segmentation of Optical Coherence Tomography Images

Algorithms ◽  
2020 ◽  
Vol 13 (3) ◽  
pp. 60 ◽  
Author(s):  
Wen Liu ◽  
Yankui Sun ◽  
Qingge Ji
Keyword(s):  
Optical Coherence Tomography ◽  
State Of The Art ◽  
Imaging Technique ◽  
Optical Coherence ◽  
Convolutional Network ◽  
Multi Scale ◽  
Segmentation Methods ◽  
Benchmark Datasets ◽  
Resolution Imaging ◽  
Layer Segmentation

Optical coherence tomography (OCT) is an optical high-resolution imaging technique for ophthalmic diagnosis. In this paper, we take advantages of multi-scale input, multi-scale side output and dual attention mechanism and present an enhanced nested U-Net architecture (MDAN-UNet), a new powerful fully convolutional network for automatic end-to-end segmentation of OCT images. We have evaluated two versions of MDAN-UNet (MDAN-UNet-16 and MDAN-UNet-32) on two publicly available benchmark datasets which are the Duke Diabetic Macular Edema (DME) dataset and the RETOUCH dataset, in comparison with other state-of-the-art segmentation methods. Our experiment demonstrates that MDAN-UNet-32 achieved the best performance, followed by MDAN-UNet-16 with smaller parameter, for multi-layer segmentation and multi-fluid segmentation respectively.

scholarly journals Learn from Syntax: Improving Pair-wise Aspect and Opinion Terms Extraction with Rich Syntactic Knowledge

2021 ◽  
Author(s):  
Shengqiong Wu ◽  
Hao Fei ◽  
Yafeng Ren ◽  
Donghong Ji ◽  
Jingye Li
Keyword(s):  
State Of The Art ◽  
Boundary Detection ◽  
High Order ◽  
Experimental Results ◽  
Convolutional Network ◽  
Syntactic Knowledge ◽  
Current State ◽  
Syntactic Features ◽  
Benchmark Datasets

In this paper, we propose to enhance the pair-wise aspect and opinion terms extraction (PAOTE) task by incorporating rich syntactic knowledge. We first build a syntax fusion encoder for encoding syntactic features, including a label-aware graph convolutional network (LAGCN) for modeling the dependency edges and labels, as well as the POS tags unifiedly, and a local-attention module encoding POS tags for better term boundary detection. During pairing, we then adopt Biaffine and Triaffine scoring for high-order aspect-opinion term pairing, in the meantime re-harnessing the syntax-enriched representations in LAGCN for syntactic-aware scoring. Experimental results on four benchmark datasets demonstrate that our model outperforms current state-of-the-art baselines, meanwhile yielding explainable predictions with syntactic knowledge.

scholarly journals Identification of hsa-miR-106a-5p as an impact agent on promotion of multiple sclerosis using multi-step data analysis

2020 ◽  
Author(s):  
Samira Rahimirad ◽  
Mohammad Navaderi ◽  
Shokoofeh Alaei ◽  
Mohammad Hossein Sanati
Keyword(s):  
Gene Expression ◽  
Multiple Sclerosis ◽  
Demyelinating Disease ◽  
Expression Patterns ◽  
Regulation Of Gene Expression ◽  
Blood Samples ◽  
Non Coding Rna ◽  
Healthy Control ◽  
Post Transcriptional Regulation ◽  
Signal Transductions

AbstractMultiple Sclerosis (MS) is a chronic, demyelinating disease in which the neuron myelin sheath is disrupted and leading to signal transductions disabilities. The evidence demonstrated that gene expression patterns and their related regulating factors are the most critical agents in Multiple Sclerosis demyelinating process. A miRNA is a small non-coding RNA which functions in post-transcriptional regulation of gene expression. Identification of specific miRNA dysregulation patterns in multiple sclerosis blood samples compared to healthy control can be used as a diagnostic and prognostic agent. Through the literature review and bioinformatics analysis, it was found that the hsa-miR-106a-5p can be considered as a significant MS pathogenic factor, which seems has an abnormal expression pattern in patients’ blood. Experimental validation using Real-Time PCR assay was carried to verifying the miR-106a-5p expression in Multiple Sclerosis and healthy control blood samples. The obtained results proved the miR-106a dysregulation in MS patients. The expression levels of miR-106a-5p were significantly down-regulated (Fold change=0.44) in patient blood samples compared to controls (p=0.059). Our study suggested that miR-106a-5p may have a biomarker potential to the diagnosis of MS patients based on its dysregulation patterns in Multiple Sclerosis blood.

scholarly journals Training Group Orthogonal Neural Networks with Privileged Information

2017 ◽  
Author(s):  
Yunpeng Chen ◽  
Xiaojie Jin ◽  
Jiashi Feng ◽  
Shuicheng Yan
Keyword(s):  
Neural Networks ◽  
State Of The Art ◽  
Training Group ◽  
Privileged Information ◽  
Training Process ◽  
Deep Convolutional Neural Networks ◽  
Absolute Value ◽  
Generalization Ability ◽  
Training Images ◽  
Benchmark Datasets

Learning rich and diverse representations is critical for the performance of deep convolutional neural networks (CNNs). In this paper, we consider how to use privileged information to promote inherent diversity of a single CNN model such that the model can learn better representations and offer stronger generalization ability. To this end, we propose a novel group orthogonal convolutional neural network (GoCNN) that learns untangled representations within each layer by exploiting provided privileged information and enhances representation diversity effectively. We take image classification as an example where image segmentation annotations are used as privileged information during the training process. Experiments on two benchmark datasets – ImageNet and PASCAL VOC – clearly demonstrate the strong generalization ability of our proposed GoCNN model. On the ImageNet dataset, GoCNN improves the performance of state-of-the-art ResNet-152 model by absolute value of 1.2% while only uses privileged information of 10% of the training images, confirming effectiveness of GoCNN on utilizing available privileged knowledge to train better CNNs.

scholarly journals MiRNAs as promising biomarkers in cancer

2019 ◽  
Vol 11 (3) ◽  
pp. 5-12
Author(s):  
Andrian A. Tarmaev ◽  
Ozal A. Beylerli
Keyword(s):  
Genomic Sequence ◽  
Human Cancer ◽  
Regulation Of Gene Expression ◽  
Metastatic Potential ◽  
Protein Coding ◽  
Tumor Invasiveness ◽  
Eukaryotic Genes ◽  
Non Coding Rna ◽  
Gene Transcripts ◽  
Post Transcriptional Regulation

By 2004 according to the resuts of the international genome sequencing, about 20,000 protein-coding genes had been analyzed which correspond to more than 2% of the total genomic sequence. The vast majority of gene transcripts are actually characterized as non-coding RNA (ncRNA) and are a cluster of RNA that do not encode functional proteins. They can be small, approximately 20 nucleotides in length, known as microRNAs (miRNAs), or transcripts longer than 200 nucleotides, defined as long non-coding RNAs (lncRNAs). miRNAs are short ncRNAs that are involved in the post-transcriptional regulation of gene expression. Discovered over 15 years ago, these molecules have been recognized as one of the main regulatory molecules in the human genome. They play an important role in all biological processes being important modulators of the eukaryotic genes expression. Focusing on transcripts that encode proteins, miRNAs influence the cellular transcript, thereby helping to determine the outcome of the cell. Aberrant miRNA expression was observed in cancer patients. Tissue concentrations of specific miRNAs have been shown to be associated with tumor invasiveness, metastatic potential, and other clinical characteristics for many types of cancer. Compared to traditional biomarkers like proteins, miRNA has several advantages that will allow them to be considered as new potential biomarkers in cancer. This review looks at biogenesis, functions, technologies used to detect miRNA, and the association of miRNA with human cancer, in particular, colorectal cancer.

scholarly journals Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-17
Author(s):  
Cuijie Zhao ◽  
Hongdong Zhao ◽  
Guozhen Wang ◽  
Hong Chen
Keyword(s):  
Hyperspectral Image ◽  
Computational Cost ◽  
Ground Truth ◽  
Hyperspectral Data ◽  
Deep Convolutional Neural Networks ◽  
Convolutional Network ◽  
Benchmark Datasets ◽  
Deep Convolutional Network ◽  
3D Cnn

At present, the classification of the hyperspectral image (HSI) based on the deep convolutional network has made great progress. Due to the high dimensionality of spectral features, limited samples of ground truth, and high nonlinearity of hyperspectral data, effective classification of HSI based on deep convolutional neural networks is still difficult. This paper proposes a novel deep convolutional network structure, namely, a hybrid depth-separable residual network, for HSI classification, called HDSRN. The HDSRN model organically combines 3D CNN, 2D CNN, multiresidual network ROR, and depth-separable convolutions to extract deeper abstract features. On the one hand, due to the addition of multiresidual structures and skip connections, this model can alleviate the problem of over fitting, help the backpropagation of gradients, and extract features more fully. On the other hand, the depth-separable convolutions are used to learn the spatial feature, which reduces the computational cost and alleviates the decline in accuracy. Extensive experiments on the popular HSI benchmark datasets show that the performance of the proposed network is better than that of the existing prevalent methods.

scholarly journals The role of microRNAs in angiogenesis

2019 ◽  
Vol 65 (4) ◽  
Author(s):  
Joanna Bujak ◽  
Patrycja Kopytko ◽  
Małgorzata Lubecka ◽  
Katarzyna Sokołowska ◽  
Maciej Tarnowski
Keyword(s):  
Regulation Of Gene Expression ◽  
Tumor Development ◽  
Main Function ◽  
Rna Molecules ◽  
Pathological Conditions ◽  
Non Coding Rna ◽  
Antiangiogenic Factors ◽  
And Migration ◽  
Post Transcriptional Regulation

Angiogenesis is the process that leads to the formation of new blood vessels. Under physiological conditions it occurs, inter alia, during corpus luteum formation and in some stages of the menstrual cycle. However, angiogenesis plays an essential role in many pathological conditions, particularly cancer. New blood vessel formation provides cancer cells with oxygen and essential nutrients, which stimulates tumor growth and facilitates its metastasis. Increasing evidence indicates that angiogenesis is regulated by microRNAs (miRNAs), which are small non-coding RNA molecules of 19–25 nucleotides. The main function of miRNAs is post-transcriptional regulation of gene expression, which controls many key biological processes, including cell proliferation, differentiation and migration. Endothelial miRNAs, known as angiomiRs, are presumably involved in tumor development and angiogenesis through regulation of pro- and antiangiogenic factors. To date, the miRNAs that stimulate angiogenesis are: miR-9, miR-27a, miR-30d, miR0-130b, miR-139, miR-146a, miR-150, miR-155, miR-200c, miR-296 and miR-558. Conversely, miRNAs that inhibit angiogenesis are: miR-145, miR-519c, miR-22, miR-20a, miR-92, miR-7b, miR-221, miR-222, miR-328 and miR-101.

scholarly journals In Silico Analysis of Osa-miR164 Gene Family in Rice (Oryza Sativa)

2021 ◽  
Vol 37 (3) ◽  
Author(s):  
Vuong Quang Tien ◽  
Nguyen Huy Duong ◽  
Dao Trong Nhan ◽  
Phan Minh Vu ◽  
Do Thi Phuc
Keyword(s):  
Oryza Sativa ◽  
Expression Pattern ◽  
Target Genes ◽  
Regulation Of Gene Expression ◽  
In Silico Analysis ◽  
Regulatory Elements ◽  
Stress Conditions ◽  
Gene Promoters ◽  
Non Coding Rna ◽  
Post Transcriptional Regulation

MicroRNA (miRNA) is a small non-coding RNA molecule containing about 22- 24 nucleotides, which functions in post-transcriptional regulation of gene expression. Previous reports have shown that miRNA plays an important role on the resistance ability of plants to adverse conditions. Rice (Oryza sativa) is a major food crop. Climate change makes the situation of salinity and drought in Vietnam worse, significantly affects rice cultivation area, leading to the decrease of the quantity and the quality of rice grains. In this research, we focused on miR164 family in rice. By using bioinformatics approach, we analyzed sequences of all osa-miR164 belonging to rice miR164 family, evaluated the expression profile of osa-miR164 under different stress conditions, predicted cis-regulatory elements on osa-miR164 gene promoters, and simultaneously predicted miR164-targeted genes and their expressions. The results showed the high conserve in mature osa-miR164 sequences but not in the precursor sequences, different expression pattern of osa-miR164 gene members under stress conditions and various cis-regulatory elements present in osa-miR164 gene promoters which may explain for diverse expression pattern of osa-miR164 genes. Some potential target genes of osa-miR164 were identified and their expressions under different stress conditions were analyzed.

scholarly journals MicroRNAs as Biomarkers for Animal Health and Welfare in Livestock

2020 ◽  
Vol 7 ◽  
Author(s):  
Silvia Miretti ◽  
Cristina Lecchi ◽  
Fabrizio Ceciliani ◽  
Mario Baratta
Keyword(s):  
Current Knowledge ◽  
Animal Health ◽  
Regulation Of Gene Expression ◽  
Cell Communication ◽  
Rna Molecules ◽  
Non Invasive ◽  
Non Coding Rna ◽  
Wide Range ◽  
Post Transcriptional Regulation ◽  
Extracellular Environment

MicroRNAs (miRNAs) are small and highly conserved non-coding RNA molecules that orchestrate a wide range of biological processes through the post-transcriptional regulation of gene expression. An intriguing aspect in identifying these molecules as biomarkers is derived from their role in cell-to-cell communication, their active secretion from cells into the extracellular environment, their high stability in body fluids, and their ease of collection. All these features confer on miRNAs the potential to become a non-invasive tool to score animal welfare. There is growing interest in the importance of miRNAs as biomarkers for assessing the welfare of livestock during metabolic, environmental, and management stress, particularly in ruminants, pigs, and poultry. This review provides an overview of the current knowledge regarding the potential use of tissue and/or circulating miRNAs as biomarkers for the assessment of the health and welfare status in these livestock species.

scholarly journals Context Modulated Dynamic Networks for Actor and Action Video Segmentation with Language Queries

2020 ◽  
Vol 34 (07) ◽  
pp. 12152-12159
Author(s):  
Hao Wang ◽  
Cheng Deng ◽  
Fan Ma ◽  
Yi Yang
Keyword(s):  
Video Segmentation ◽  
State Of The Art ◽  
Dynamic Networks ◽  
Specific Region ◽  
Visual Features ◽  
Convolutional Network ◽  
Fine Grained ◽  
Convolutional Networks ◽  
Benchmark Datasets ◽  
Context Features

Actor and action video segmentation with language queries aims to segment out the expression referred objects in the video. This process requires comprehensive language reasoning and fine-grained video understanding. Previous methods mainly leverage dynamic convolutional networks to match visual and semantic representations. However, the dynamic convolution neglects spatial context when processing each region in the frame and is thus challenging to segment similar objects in the complex scenarios. To address such limitation, we construct a context modulated dynamic convolutional network. Specifically, we propose a context modulated dynamic convolutional operation in the proposed framework. The kernels for the specific region are generated from both language sentences and surrounding context features. Moreover, we devise a temporal encoder to incorporate motions into the visual features to further match the query descriptions. Extensive experiments on two benchmark datasets, Actor-Action Dataset Sentences (A2D Sentences) and J-HMDB Sentences, demonstrate that our proposed approach notably outperforms state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document