Publications by authors named "Kil To Chong"

30 Publications

  • Page 1 of 1

Identifying DNA N4-methylcytosine sites in the rosaceae genome with a deep learning model relying on distributed feature representation.

Comput Struct Biotechnol J 2021 19;19:1612-1619. Epub 2021 Mar 19.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, South Korea.

DNA N4-methylcytosine (4mC), an epigenetic modification found in prokaryotic and eukaryotic species, is involved in numerous biological functions, including host defense, transcription regulation, gene expression, and DNA replication. To identify 4mC sites, previous computational studies mostly focused on finding hand-crafted features. This area of research, therefore, would benefit from the development of a computational approach that relies on automatic feature selection to identify relevant sites. We here report 4mC-w2vec, a computational method that learned automatic feature discrimination in the genomes, especially in and , based on distributed feature representation and through the word embedding technique 'word2vec'. While a few bioinformatics tools are currently employed to identify 4mC sites in these , their prediction performance is inadequate. Our system processed 4mC and non-4mC sites through a word embedding process, including sub-word information of its biological words through k-mer, which then served as features that were fed into a double layer of convolutional neural network (CNN) to classify whether the sample sequences contained 4mCs or non-4mCs sites. Our tool demonstrated performance superior to current tools that use the same genomic datasets. Additionally, 4mC-w2vec is effective for balanced and imbalanced class datasets alike, and the online web-server is currently available at: http://nsclbio.jbnu.ac.kr/tools/4mC-w2vec/.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.csbj.2021.03.015DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8042287PMC
March 2021

4mCPred-CNN-Prediction of DNA N4-Methylcytosine in the Mouse Genome Using a Convolutional Neural Network.

Genes (Basel) 2021 Feb 20;12(2). Epub 2021 Feb 20.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, Korea.

Among DNA modifications, N4-methylcytosine (4mC) is one of the most significant ones, and it is linked to the development of cell proliferation and gene expression. To know different its biological functions, the accurate detection of 4mC sites is required. Although we have several techniques for the prediction of 4mC sites in different genomes based on both machine learning (ML) and convolutional neural networks (CNNs), there is no CNN-based tool for the identification of 4mC sites in the mouse genome. In this article, a CNN-based model named 4mCPred-CNN was developed to classify 4mC locations in the mouse genome. Until now, we had only two ML-based models for this purpose; they utilized several feature encoding schemes, and thus still had a lot of space available to improve the prediction accuracy. Utilizing only a single feature encoding scheme-one-hot encoding-we outperformed both of the previous ML-based techniques. In a ten-fold validation test, the proposed model, 4mCPred-CNN, achieved an accuracy of 85.71% and Matthews correlation coefficient (MCC) of 0.717. On an independent dataset, the achieved accuracy was 87.50% with an MCC value of 0.750. The attained results exhibit that the proposed model can be of great use for researchers in the fields of biology and bioinformatics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/genes12020296DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7924022PMC
February 2021

BrainSeg-Net: Brain Tumor MR Image Segmentation via Enhanced Encoder-Decoder Network.

Diagnostics (Basel) 2021 Jan 25;11(2). Epub 2021 Jan 25.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, Korea.

Efficient segmentation of Magnetic Resonance (MR) brain tumor images is of the utmost value for the diagnosis of tumor region. In recent years, advancement in the field of neural networks has been used to refine the segmentation performance of brain tumor sub-regions. The brain tumor segmentation has proven to be a complicated task even for neural networks, due to the small-scale tumor regions. These small-scale tumor regions are unable to be identified, the reason being their tiny size and the huge difference between area occupancy by different tumor classes. In previous state-of-the-art neural network models, the biggest problem was that the location information along with spatial details gets lost in deeper layers. To address these problems, we have proposed an encoder-decoder based model named BrainSeg-Net. The Feature Enhancer (FE) block is incorporated into the BrainSeg-Net architecture which extracts the middle-level features from low-level features from the shallow layers and shares them with the dense layers. This feature aggregation helps to achieve better performance of tumor identification. To address the problem associated with imbalance class, we have used a custom-designed loss function. For evaluation of BrainSeg-Net architecture, three benchmark datasets are utilized: BraTS2017, BraTS 2018, and BraTS 2019. Segmentation of Enhancing Core (EC), Whole Tumor (WT), and Tumor Core (TC) is carried out. The proposed architecture have exhibited good improvement when compared with existing baseline and state-of-the-art techniques. The MR brain tumor segmentation by BrainSeg-Net uses enhanced location and spatial features, which performs better than the existing plethora of brain MR image segmentation approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/diagnostics11020169DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7911842PMC
January 2021

Prediction of kinase inhibitors binding modes with machine learning and reduced descriptor sets.

Sci Rep 2021 Jan 12;11(1):706. Epub 2021 Jan 12.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju, 54896, South Korea.

Protein kinases are receiving wide research interest, from drug perspective, due to their important roles in human body. Available kinase-inhibitor data, including crystallized structures, revealed many details about the mechanism of inhibition and binding modes. The understanding and analysis of these binding modes are expected to support the discovery of kinase-targeting drugs. The huge amounts of data made it possible to utilize computational techniques, including machine learning, to help in the discovery of kinase-targeting drugs. Machine learning gave reasonable predictions when applied to differentiate between the binding modes of kinase inhibitors, promoting a wider application in that domain. In this study, we applied machine learning supported by feature selection techniques to classify kinase inhibitors according to their binding modes. We represented inhibitors as a large number of molecular descriptors, as features, and systematically reduced these features in a multi-step manner while trying to attain high classification accuracy. Our predictive models could satisfy both goals by achieving high accuracy while utilizing at most 5% of the modeling features. The models could differentiate between binding mode types with MCC values between 0.67 and 0.92, and balanced accuracy values between 0.78 and 0.97 for independent test sets.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-020-80758-4DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7804204PMC
January 2021

DNA sequences performs as natural language processing by exploiting deep learning algorithm for the identification of N4-methylcytosine.

Sci Rep 2021 Jan 8;11(1):212. Epub 2021 Jan 8.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju, 54896, South Korea.

N4-methylcytosine is a biochemical alteration of DNA that affects the genetic operations without modifying the DNA nucleotides such as gene expression, genomic imprinting, chromosome stability, and the development of the cell. In the proposed work, a computational model, 4mCNLP-Deep, used the word embedding approach as a vector formulation by exploiting deep learning based CNN algorithm to predict 4mC and non-4mC sites on the C.elegans genome dataset. Diversity of ranges employed for the experimental such as corpus k-mer and k-fold cross-validation to obtain the prevailing capabilities. The 4mCNLP-Deep outperform from the state-of-the-art predictor by achieving the results in five evaluation metrics by following; Accuracy (ACC) as 0.9354, Mathew's correlation coefficient (MCC) as 0.8608, Specificity (Sp) as 0.89.96, Sensitivity (Sn) as 0.9563, and Area under curve (AUC) as 0.9731 by using 3-mer corpus word2vec and 3-fold cross-validation and attained the increment of 1.1%, 0.6%, 0.58%, 0.77%, and 4.89%, respectively. At last, we developed the online webserver http://nsclbio.jbnu.ac.kr/tools/4mCNLP-Deep/ , for the experimental researchers to get the results easily.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-020-80430-xDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7794489PMC
January 2021

pcPromoter-CNN: A CNN-Based Prediction and Classification of Promoters.

Genes (Basel) 2020 Dec 21;11(12). Epub 2020 Dec 21.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, Korea.

A promoter is a small region within the DNA structure that has an important role in initiating transcription of a specific gene in the genome. Different types of promoters are recognized by their different functions. Due to the importance of promoter functions, computational tools for the prediction and classification of a promoter are highly desired. Promoters resemble each other; therefore, their precise classification is an important challenge. In this study, we propose a convolutional neural network (CNN)-based tool, the pcPromoter-CNN, for application in the prediction of promotors and their classification into subclasses σ70, σ54, σ38, σ32, σ28 and σ24. This CNN-based tool uses a one-hot encoding scheme for promoter classification. The tools architecture was trained and tested on a benchmark dataset. To evaluate its classification performance, we used four evaluation metrics. The model exhibited notable improvement over that of existing state-of-the-art tools.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/genes11121529DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7767505PMC
December 2020

Carotid Artery Plaque Identification and Display System (MRI-CAPIDS) Using Opensource Tools.

Diagnostics (Basel) 2020 Dec 21;10(12). Epub 2020 Dec 21.

Electronic Engineering Department, Jeonbuk National University, Jeonju 54896, Korea.

Magnetic resonance imaging (MRI) represents one modality in atherosclerosis risk assessment, by permitting the classification of carotid plaques into either high- or low-risk lesions. Although MRI is generally used for observing the impact of atherosclerosis on vessel lumens, it can also show both the size and composition of itself, as well as plaque information, thereby providing information beyond that of simple stenosis. Software systems are a valuable aid in carotid artery stenosis assessment wherein commercial software is readily available but is not accessible to all practitioners because of its often high cost. This study focuses on the development of a software system designed entirely for registration, marking, and 3D visualization of the wall and lumen, using freely available open-source tools and libraries. It was designed to be free from "feature bloat" and avoid "feature-creep." The image loading and display module of the modified QDCM library was improved by a minimum of 10,000%. A Bezier function was used in order to smoothen the curve of the polygon (referring to the shape formed by the marked points) by interpolating additional points between the marked points. This smoother curve led to a smoother 3D view of the lumen and wall.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/diagnostics10121111DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7767364PMC
December 2020

XG-ac4C: identification of N4-acetylcytidine (ac4C) in mRNA using eXtreme gradient boosting with electron-ion interaction pseudopotentials.

Sci Rep 2020 12 1;10(1):20942. Epub 2020 Dec 1.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju, 54896, South Korea.

N4-acetylcytidine (ac4C) is a post-transcriptional modification in mRNA which plays a major role in the stability and regulation of mRNA translation. The working mechanism of ac4C modification in mRNA is still unclear and traditional laboratory experiments are time-consuming and expensive. Therefore, we propose an XG-ac4C machine learning model based on the eXtreme Gradient Boost classifier for the identification of ac4C sites. The XG-ac4C model uses a combination of electron-ion interaction pseudopotentials and electron-ion interaction pseudopotentials of trinucleotide of the nucleotides in ac4C sites. Moreover, Shapley additive explanations and local interpretable model-agnostic explanations are applied to understand the importance of features and their contribution to the final prediction outcome. The obtained results demonstrate that XG-ac4C outperforms existing state-of-the-art methods. In more detail, the proposed model improves the area under the precision-recall curve by 9.4% and 9.6% in cross-validation and independent tests, respectively. Finally, a user-friendly web server based on the proposed model for ac4C site identification is made freely available at http://nsclbio.jbnu.ac.kr/tools/xgac4c/ .
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-020-77824-2DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7708984PMC
December 2020

Early and accurate detection and diagnosis of heart disease using intelligent computational model.

Sci Rep 2020 11 12;10(1):19747. Epub 2020 Nov 12.

Department of Electronic and Information Engineering, Jeonbuk National University, Jeonju, 54896, South Korea.

Heart disease is a fatal human disease, rapidly increases globally in both developed and undeveloped countries and consequently, causes death. Normally, in this disease, the heart fails to supply a sufficient amount of blood to other parts of the body in order to accomplish their normal functionalities. Early and on-time diagnosing of this problem is very essential for preventing patients from more damage and saving their lives. Among the conventional invasive-based techniques, angiography is considered to be the most well-known technique for diagnosing heart problems but it has some limitations. On the other hand, the non-invasive based methods, like intelligent learning-based computational techniques are found more upright and effectual for the heart disease diagnosis. Here, an intelligent computational predictive system is introduced for the identification and diagnosis of cardiac disease. In this study, various machine learning classification algorithms are investigated. In order to remove irrelevant and noisy data from extracted feature space, four distinct feature selection algorithms are applied and the results of each feature selection algorithm along with classifiers are analyzed. Several performance metrics namely: accuracy, sensitivity, specificity, AUC, F1-score, MCC, and ROC curve are used to observe the effectiveness and strength of the developed model. The classification rates of the developed system are examined on both full and optimal feature spaces, consequently, the performance of the developed model is boosted in case of high variated optimal feature space. In addition, P-value and Chi-square are also computed for the ET classifier along with each feature selection technique. It is anticipated that the proposed system will be useful and helpful for the physician to diagnose heart disease accurately and effectively.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-020-76635-9DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7665174PMC
November 2020

i6mA-stack: A stacking ensemble-based computational prediction of DNA N6-methyladenine (6mA) sites in the Rosaceae genome.

Genomics 2021 Jan 1;113(1 Pt 2):582-592. Epub 2020 Oct 1.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, South Korea; Advanced Electronics and Information Research Center, Jeonbuk National University, Jeonju 54896, South Korea. Electronic address:

DNA N6-methyladenine (6 mA) is an epigenetic modification that plays a vital role in a variety of cellular processes in both eukaryotes and prokaryotes. Accurate information of 6 mA sites in the Rosaceae genome may assist in understanding genomic 6 mA distributions and various biological functions such as epigenetic inheritance. Various studies have shown the possibility of identifying 6 mA sites through experiments, but the procedures are time-consuming and costly. To overcome the drawbacks of experimental methods, we propose an accurate computational paradigm based on a machine learning (ML) technique to identify 6 mA sites in Rosa chinensis (R.chinensis) and Fragaria vesca (F.vesca). To improve the performance of the proposed model and to avoid overfitting, a recursive feature elimination with cross-validation (RFECV) strategy is used to extract the optimal number of features (ONF) subset from five different DNA sequence encoding schemes, i.e., Binary Encoding (BE), Ring-Function-Hydrogen-Chemical Properties (RFHC), Electron-Ion-Interaction Pseudo Potentials of Nucleotides (EIIP), Dinucleotide Physicochemical Properties (DPCP), and Trinucleotide Physicochemical Properties (TPCP). Subsequently, we use the ONF subset to train a double layers of ML-based stacking model to create a bioinformatics tool named 'i6mA-stack'. This tool outperforms its peer tool in general and is currently available at http://nsclbio.jbnu.ac.kr/tools/i6mA-stack/.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ygeno.2020.09.054DOI Listing
January 2021

ncRDeep: Non-coding RNA classification with convolutional neural network.

Comput Biol Chem 2020 Oct 27;88:107364. Epub 2020 Aug 27.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, South Korea; Advanced Electronics and Information Research Center, Jeonbuk National University, Jeonju 54896, South Korea. Electronic address:

A non-coding RNA (ncRNA) is a kind of RNA that is not converted into protein, however, it is involved in many biological processes, diseases, and cancers. Numerous ncRNAs have been identified and classified with high throughput sequencing technology. Hence, accurate ncRNAs class prediction is important and necessary for further study of their functions. Several computation techniques have been employed to predict the class of ncRNAs. Recent classification methods used the secondary structure as their primary input. However, the computational tools of RNA secondary structure are not accurate enough which affects the final performance of ncRNAs predictors. In this paper, we propose a simple yet efficient method, called ncRDeep, for ncRNAs prediction. It uses a simple convolutional neural network and RNA sequence information only. The ncRDeep was evaluated on benchmark datasets and the comparison results showed that the ncRDeep outperforms the state-of-the-art methods significantly. More specifically, the average accuracy was improved by 8.32%. Finally, we built a freely accessible web server for the developed tool ncRDeep at http://home.jbnu.ac.kr/NSCL/ncRDeep.htm.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.compbiolchem.2020.107364DOI Listing
October 2020

DNA6mA-MINT: DNA-6mA Modification Identification Neural Tool.

Genes (Basel) 2020 08 5;11(8). Epub 2020 Aug 5.

Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, Korea.

DNA N6-methyladenine (6mA) is part of numerous biological processes including DNA repair, DNA replication, and DNA transcription. The 6mA modification sites hold a great impact when their biological function is under consideration. Research in biochemical experiments for this purpose is carried out and they have demonstrated good results. However, they proved not to be a practical solution when accessed under cost and time parameters. This led researchers to develop computational models to fulfill the requirement of modification identification. In consensus, we have developed a computational model recommended by Chou's 5-steps rule. The Neural Network (NN) model uses convolution layers to extract the high-level features from the encoded binary sequence. These extracted features were given an optimal interpretation by using a Long Short-Term Memory (LSTM) layer. The proposed architecture showed higher performance compared to state-of-the-art techniques. The proposed model is evaluated on , Rice, and "Combined-species" genomes with 5- and 10-fold cross-validation. Further, with access to a user-friendly web server, publicly available can be accessed freely.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/genes11080898DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7463462PMC
August 2020

DNC4mC-Deep: Identification and Analysis of DNA N4-Methylcytosine Sites Based on Different Encoding Schemes By Using Deep Learning.

Cells 2020 07 22;9(8). Epub 2020 Jul 22.

Department of Electronics Engineering, Jeonbuk National University, Jeonju 54896, Korea.

N4-methylcytosine as one kind of modification of DNA has a critical role which alters genetic performance such as protein interactions, conformation, stability in DNA as well as the regulation of gene expression same cell developmental and genomic imprinting. Some different 4mC site identifiers have been proposed for various species. Herein, we proposed a computational model, DNC4mC-Deep, including six encoding techniques plus a deep learning model to predict 4mC sites in the genome of , , and Cross-species dataset. It was demonstrated by the 10-fold cross-validation test to get superior performance. The DNC4mC-Deep obtained 0.829 and 0.929 of MCC on and training dataset, respectively, and 0.814 on cross-species. This means the proposed method outperforms the state-of-the-art predictors at least 0.284 and 0.265 on and training dataset in turn. Furthermore, the DNC4mC-Deep achieved 0.635 and 0.565 of MCC on and independent dataset, respectively, and 0.562 on cross-species which shows it can achieve the best performance to predict 4mC sites as compared to the state-of-the-art predictor.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/cells9081756DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7465362PMC
July 2020

Prediction of N6-methyladenosine sites using convolution neural network model based on distributed feature representations.

Neural Netw 2020 Sep 2;129:385-391. Epub 2020 Jun 2.

Department of Electronics and Information Engineering, Chonbuk National University, Jeonju 54896, South Korea; Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, South Korea. Electronic address:

N-methyladenosine (mA) is a well-studied and most common interior messenger RNA (mRNA) modification that plays an important function in cell development. NA is found in all kingdoms​ of life and many other cellular processes such as RNA splicing, immune tolerance, regulatory functions, RNA processing, and cancer. Despite the crucial role of mA in cells, it was targeted computationally, but unfortunately, the obtained results were unsatisfactory. It is imperative to develop an efficient computational model that can truly represent mA sites. In this regard, an intelligent and highly discriminative computational model namely: m6A-word2vec is introduced for the discrimination of mA sites. Here, a concept of natural language processing in the form of word2vec is used to represent the motif of the target class automatically. These motifs (numerical descriptors) are automatically targeted from the human genome without any clear definition. Further, the extracted feature space is then forwarded to the convolution neural network model as input for prediction. The developed computational model obtained 83.17%, 92.69%, and 90.50% accuracy for benchmark datasets S, S, and S, respectively, using a 10-fold cross-validation test. The predictive outcomes validate that the developed intelligent computational model showed better performance compared to existing computational models. It is thus greatly estimated that the introduced computational model "m6A-word2vec" may be a supportive and practical tool for elementary and pharmaceutical research such as in drug design along with academia.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neunet.2020.05.027DOI Listing
September 2020

iMethyl-Deep: N6 Methyladenosine Identification of Yeast Genome with Automatic Feature Extraction Technique by Using Deep Learning Algorithm.

Genes (Basel) 2020 05 9;11(5). Epub 2020 May 9.

Advanced Electronics and Information Research Center, Jeonbuk National University, Jeonju 54896, Korea.

One of the most common and well studied post-transcription modifications in RNAs is N6-methyladenosine (m6A) which has been involved with a wide range of biological processes. Over the past decades, N6-methyladenosine produced some positive consequences through the high-throughput laboratory techniques but still, these lab processes are time consuming and costly. Diverse computational methods have been proposed to identify m6A sites accurately. In this paper, we proposed a computational model named iMethyl-deep to identify m6A on two benchmark datasets M6A2614 and M6A6540 by using single nucleotide resolution to convert RNA sequence into a high quality feature representation. The iMethyl-deep obtained 89.19% and 87.44% of accuracy on M6A2614 and M6A6540 respectively which show that our proposed method outperforms the state-of-the-art predictors, at least 8.44%, 8.96%, 8.69% and 0.173 on M6A2614 and 15.47%, 28.52%, 25.54 and 0.5 on M6A6540 higher in terms of four metrics Sp, Sn, ACC and MCC respectively. Meanwhile, M6A6540 dataset never used to train a model.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/genes11050529DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7288457PMC
May 2020

Improving the Quantification of DNA Sequences Using Evolutionary Information Based on Deep Learning.

Cells 2019 12 14;8(12). Epub 2019 Dec 14.

Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, Korea.

It is known that over 98% of the human genome is non-coding, and 93% of disease associated variants are located in these regions. Therefore, understanding the function of these regions is important. However, this task is challenging as most of these regions are not well understood in terms of their functions. In this paper, we introduce a novel computational model based on deep neural networks, called DQDNN, for quantifying the function of non-coding DNA regions. This model combines convolution layers for capturing regularity motifs at multiple scales and recurrent layers for capturing long term dependencies between the captured motifs. In addition, we show that integrating evolutionary information with raw genomic sequences improves the performance of the predictor significantly. The proposed model outperforms the state-of-the-art ones using raw genomics sequences only and also by integrating evolutionary information with raw genomics sequences. More specifically, the proposed model improves 96.9% and 98% of the targets in terms of area under the receiver operating characteristic curve and the precision-recall curve, respectively. In addition, the proposed model improved the prioritization of functional variants of expression quantitative trait loci (eQTLs) compared with the state-of-the-art models.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/cells8121635DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6952993PMC
December 2019

Coordination optimization control of DC component and harmonics for grid-connected PV inverters.

ISA Trans 2020 Apr 25;99:432-444. Epub 2019 Sep 25.

Department of Electronic Engineering, Chonbuk National University, Jeonju, 54896, South Korea. Electronic address:

Grid-connected inverters (GCIs) have been extensively adopted in distributed renewable energy systems. However, due to the asymmetrical gate-driving signals, imparities of the power semiconductors, and unbalanced grid voltage, etc., the grid current contains DC components which are detrimental to the grid. Furthermore, due to the nonlinear loads in the grid, the grid current may also contain harmonics. As a result, power quality of the grid current is degraded. IEEE 1547-2003 has specified the maximum DC component and harmonics contents of the grid current. In this paper, a repetitive controller (RC) for DC component compensation is put forward at first, then, a Discrete-Fourier-Transformation (DFT)-RC controller is presented for harmonic elimination. Considering DC components and harmonics may coexist in GCIs in utility, a coordination optimization control scheme utilizing Weight Factor Distribution Optimization Method (WFDOM) is introduced. The proposed method is realized by adaptively allocating the weight factors for DC suppression controller and harmonic elimination controller according to their relative deviations. Simulation and experimental results show the feasibility and correctness of the proposed method. The proposed method can applied to high power quality PV power generation system.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.isatra.2019.09.013DOI Listing
April 2020

Identification of prokaryotic promoters and their strength by integrating heterogeneous features.

Genomics 2020 03 19;112(2):1396-1403. Epub 2019 Aug 19.

Department of Electronics and Information Engineering, Chonbuk National University, Jeonju 54896, South Korea; Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, South Korea. Electronic address:

The promoter is a regulatory DNA region and important for gene transcriptional regulation. It is located near the transcription start site (TSS) upstream of the corresponding gene. In the post-genomics era, the availability of data makes it possible to build computational models for robustly detecting the promoters as these models are expected to be helpful for academia and drug discovery. Until recently, developed models focused only on discriminating the sequences into promoter and non-promoter. However, promoter predictors can be further improved by considering weak and strong promoter classification. In this work, we introduce a hybrid model, named iPSW(PseDNC-DL), for identification of prokaryotic promoters and their strength. It combines a convolutional neural network with a pseudo-di-nucleotide composition (PseDNC). The proposed model iPSW(PseDNC-DL) has been evaluated on the benchmark datasets and outperformed the current state-of-the-art models in both tasks namely promoter identification and promoter strength identification. The developed tool iPSW(PseDNC-DL) has been constructed in a web server and made freely available at https://home.jbnu.ac.kr/NSCL/PseDNC-DL.htm.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ygeno.2019.08.009DOI Listing
March 2020

Deep Splicing Code: Classifying Alternative Splicing Events Using Deep Learning.

Genes (Basel) 2019 08 1;10(8). Epub 2019 Aug 1.

Department of Electronics and Information Engineering, Chonbuk National University, Jeonju 54896, Korea.

Alternative splicing (AS) is the process of combining different parts of the pre-mRNA to produce diverse transcripts and eventually different protein products from a single gene. In computational biology field, researchers try to understand AS behavior and regulation using computational models known as "Splicing Codes". The final goal of these algorithms is to make an in-silico prediction of AS outcome from genomic sequence. Here, we develop a deep learning approach, called Deep Splicing Code (DSC), for categorizing the well-studied classes of AS namely alternatively skipped exons, alternative 5ss, alternative 3ss, and constitutively spliced exons based only on the sequence of the exon junctions. The proposed approach significantly improves the prediction and the obtained results reveal that constitutive exons have distinguishable local characteristics from alternatively spliced exons. Using the motif visualization technique, we show that the trained models learned to search for competitive alternative splice sites as well as motifs of important splicing factors with high precision. Thus, the proposed approach greatly expands the opportunities to improve alternative splicing modeling. In addition, a web-server for AS events prediction has been developed based on the proposed method and made available at https://home.jbnu.ac.kr/NSCL/dsc.htm.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/genes10080587DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6722613PMC
August 2019

iPseU-CNN: Identifying RNA Pseudouridine Sites Using Convolutional Neural Networks.

Mol Ther Nucleic Acids 2019 Jun 11;16:463-470. Epub 2019 Apr 11.

Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, South Korea. Electronic address:

Pseudouridine is the most prevalent RNA modification and has been found in both eukaryotes and prokaryotes. Currently, pseudouridine has been demonstrated in several kinds of RNAs, such as small nuclear RNA, rRNA, tRNA, mRNA, and small nucleolar RNA. Therefore, its significance to academic research and drug development is understandable. Through biochemical experiments, the pseudouridine site identification has produced good outcomes, but these lab exploratory methods and biochemical processes are expensive and time consuming. Therefore, it is important to introduce efficient methods for identification of pseudouridine sites. In this study, an intelligent method for pseudouridine sites using the deep-learning approach was developed. The proposed prediction model is called iPseU-CNN (identifying pseudouridine by convolutional neural networks). The existing methods used handcrafted features and machine-learning approaches to identify pseudouridine sites. However, the proposed predictor extracts the features of the pseudouridine sites automatically using a convolution neural network model. The iPseU-CNN model yields better outcomes than the current state-of-the-art models in all evaluation parameters. It is thus highly projected that the iPseU-CNN predictor will become a helpful tool for academic research on pseudouridine site prediction of RNA, as well as in drug discovery.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.omtn.2019.03.010DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6488737PMC
June 2019

DeePromoter: Robust Promoter Predictor Using Deep Learning.

Front Genet 2019 5;10:286. Epub 2019 Apr 5.

Advanced Research Center of Information and Electronics Engineering, Chonbuk National University, Jeonju, South Korea.

The promoter region is located near the transcription start sites and regulates transcription initiation of the gene by controlling the binding of RNA polymerase. Thus, promoter region recognition is an important area of interest in the field of bioinformatics. Numerous tools for promoter prediction were proposed. However, the reliability of these tools still needs to be improved. In this work, we propose a robust deep learning model, called DeePromoter, to analyze the characteristics of the short eukaryotic promoter sequences, and accurately recognize the human and mouse promoter sequences. DeePromoter combines a convolutional neural network (CNN) and a long short-term memory (LSTM). Additionally, instead of using non-promoter regions of the genome as a negative set, we derive a more challenging negative set from the promoter sequences. The proposed negative set reconstruction method improves the discrimination ability and significantly reduces the number of false positive predictions. Consequently, DeePromoter outperforms the previously proposed promoter prediction tools. In addition, a web-server for promoter prediction is developed based on the proposed methods and made available at https://home.jbnu.ac.kr/NSCL/deepromoter.htm.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fgene.2019.00286DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6460014PMC
April 2019

iRNA-PseKNC(2methyl): Identify RNA 2'-O-methylation sites by convolution neural network and Chou's pseudo components.

J Theor Biol 2019 03 24;465:1-6. Epub 2018 Dec 24.

Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, South Korea. Electronic address:

The 2'-O-methylation transferase is involved in the process of 2'-O-methylation. In catalytic processes, the 2-hydroxy group of the ribose moiety of a nucleotide accept a methyl group. This methylation process is a post-transcriptional modification, which occurs in various cellular RNAs and plays a vital role in regulation of gene expressions at the post-transcriptional level. Through biochemical experiments 2'-O-methylation sites produce good results but these biochemical process and exploratory techniques are very expensive. Thus, it is required to develop a computational method to identify 2'-O-methylation sites. In this work, we proposed a simple and precise convolution neural network method namely: iRNA-PseKNC(2methyl) to identify 2'-O-methylation sites. The existing techniques use handcrafted features, while the proposed method automatically extracts the features of 2'-O-methylation using the proposed convolution neural network model. The proposed prediction iRNA-PseKNC(2methyl) method obtained 98.27% of accuracy, 96.29% of sensitivity, 100% of specificity, and 0.965 of MCC on Home sapiens dataset. The reported outcomes present that our proposed method obtained better outcomes than existing method in terms of all evaluation parameters. These outcomes show that iRNA-PseKNC(2methyl) method might be beneficial for the academic research and drug design.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jtbi.2018.12.034DOI Listing
March 2019

An intelligent dc current minimization method for transformerless grid-connected photovoltaic inverters.

ISA Trans 2019 May 11;88:268-279. Epub 2018 Dec 11.

School of Electrical Engineering, Chonbuk National University, Jonjue, Republic of Korea. Electronic address:

Due to the scaling and zero-drift of current sensor errors, unbalanced grid voltages, tolerance of power switching devices, and asymmetry of PWM gate driving pulses, transformerless grid-connected inverters usually have certain amount of dc components injected to the ac grid. Therefore, power quality of the grid is degraded. Many efforts, such as using blocking capacitors, a dc current feedback control method, and a voltage dc component feedback control method, etc., have been introduced to minimize the dc injection. This paper proposes an intelligent control strategy of dc current injection suppression to the grid by utilizing adaptive-back-propagation (ABP) neural network PID controller. The performance of the proposed scheme is evaluated and compared with the traditional and existing method. Finally, the control scheme is verified on a 2-kW three-phase grid-connected inverter in the laboratory.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.isatra.2018.12.005DOI Listing
May 2019

Design of feedforward and feedback position control for passive bilateral teleoperation with delays.

ISA Trans 2019 Feb 20;85:200-213. Epub 2018 Oct 20.

Electronic Engineering Department, Chonbuk National University, Jeonju City, South Korea. Electronic address:

Bilateral teleoperation systems connected to computer networks such as the internet must be able to operate with varying time delays since such systems can easily become unstable. A passivity concept has been used as the framework to solve the stability problem in the bilateral control of teleoperation systems. Passivity and tracking performance are recovered using a control architecture that incorporates time varying gains into the transmission path, feedforward, and feedback position control. The proposed architecture has an inner component that can accommodate any configuration but still remain stable and passive even with varying time delay. The simulation results for a single degree of freedom master/slave system demonstrate the performance of the proposed control architecture.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.isatra.2018.10.006DOI Listing
February 2019

Object Detection in Very High-Resolution Aerial Images Using One-Stage Densely Connected Feature Pyramid Network.

Sensors (Basel) 2018 Oct 6;18(10). Epub 2018 Oct 6.

Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, Korea.

Object detection in very high-resolution (VHR) aerial images is an essential step for a wide range of applications such as military applications, urban planning, and environmental management. Still, it is a challenging task due to the different scales and appearances of the objects. On the other hand, object detection task in VHR aerial images has improved remarkably in recent years due to the achieved advances in convolution neural networks (CNN). Most of the proposed methods depend on a two-stage approach, namely: a region proposal stage and a classification stage such as Faster R-CNN. Even though two-stage approaches outperform the traditional methods, their optimization is not easy and they are not suitable for real-time applications. In this paper, a uniform one-stage model for object detection in VHR aerial images has been proposed. In order to tackle the challenge of different scales, a densely connected feature pyramid network has been proposed by which high-level multi-scale semantic feature maps with high-quality information are prepared for object detection. This work has been evaluated on two publicly available datasets and outperformed the current state-of-the-art results on both in terms of mean average precision (mAP) and computation time.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/s18103341DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6210269PMC
October 2018

Nitrogen-Doped Graphene-Encapsulated Nickel Cobalt Nitride as a Highly Sensitive and Selective Electrode for Glucose and Hydrogen Peroxide Sensing Applications.

ACS Appl Mater Interfaces 2018 Oct 12;10(42):35847-35858. Epub 2018 Oct 12.

To explore a natural nonenzymatic electrode catalyst for highly sensitive and selective molecular detection for targeting biomolecules is a very challenging task. Metal nitrides have attracted huge interest as promising electrodes for glucose and hydrogen peroxide (HO) sensing applications due to their exceptional redox properties, superior electrical conductivity, and superb mechanical strength. However, the deprived electrochemical stability extremely limits the commercialization opportunities. Herein, novel nitrogen-doped graphene-encapsulated nickel cobalt nitride (Ni CoN/NG) core-shell nanostructures with a controllable molar ratio of Ni/Co are successfully fabricated and employed as highly sensitive and selective electrodes for glucose and HO sensing applications. The highly sensitive and selective properties of the optimized core-shell NiCoN/NG electrode are because of the high synergistic effect of the NiCoN core and the NG shell, as evidenced by a superior glucose sensing performance with a short response time of <3 s, a wide linear range from 2.008 μM to 7.15 mM, an excellent sensitivity of 1803 μA mM cm, and a low detection limit of 50 nM (S/N = 3). Furthermore, the core-shell NiCoN/NG electrode shows excellent HO sensing performances with a short response time of ∼3 s, a wide detection range of 200 nM to 3.4985 mM, a high sensitivity of 2848.73 μA mM cm, and ultra-low limit detection of 200 nM (S/N = 3). The NiCoN/NG sensor can also be employed for glucose and HO detection in human blood serum, promising its application toward the determination of glucose and HO in real samples.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acsami.8b15069DOI Listing
October 2018

An Arcak-type state estimation design for time-delayed static neural networks with leakage term based on unified criteria.

Neural Netw 2018 Oct 5;106:110-126. Epub 2018 Jul 5.

School of Mathematics, and Research Center for Complex Systems and Network Sciences, Southeast University, Nanjing, 210 096, China. Electronic address:

The issue of unified dissipativity-based Arcak-type state estimator design for delayed static neural networks (SNNs) with leakage term and noise distraction was considered here. An Arcak-type state observer, which is compact than the usually used Luenberger-type state estimator, is selected to implement the subject of a unified dissipativity performance of SNNs. This paper primarily concentrates on the issue of Arcak-type state estimator of delayed SNNs involving leakage delay. The first attempt is made to tackle the Arcak-type state estimator of SNNs with time delay in leakage term in this paper based on the unified criteria, by constructing a novel Lyapunov functional together with newly improved integral inequalities. As a result, a novel unified state estimation criterion is launched in the form of linear matrix inequalities (LMIs) and put forward to justify the dynamics of error system is extended dissipative with the influence of leakage term and estimator gain matrices K¯ and K¯. Finally, an interesting simulation study is ultimately explored to show the performance of the established unified dissipativity-based theoretical results, in which, comparison results are also made together with recent works as a special case.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neunet.2018.06.015DOI Listing
October 2018

A Real-Time Marker-Based Visual Sensor Based on a FPGA and a Soft Core Processor.

Sensors (Basel) 2016 Dec 15;16(12). Epub 2016 Dec 15.

Advanced Electronics and Information Research Center, Chonbuk National University, Jeonju 54896, Korea.

This paper introduces a real-time marker-based visual sensor architecture for mobile robot localization and navigation. A hardware acceleration architecture for post video processing system was implemented on a field-programmable gate array (FPGA). The pose calculation algorithm was implemented in a System on Chip (SoC) with an Altera Nios II soft-core processor. For every frame, single pass image segmentation and Feature Accelerated Segment Test (FAST) corner detection were used for extracting the predefined markers with known geometries in FPGA. Coplanar PosIT algorithm was implemented on the Nios II soft-core processor supplied with floating point hardware for accelerating floating point operations. Trigonometric functions have been approximated using Taylor series and cubic approximation using Lagrange polynomials. Inverse square root method has been implemented for approximating square root computations. Real time results have been achieved and pixel streams have been processed on the fly without any need to buffer the input frame for further implementation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/s16122139DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5191119PMC
December 2016

Analysis of dead zone sources in a closed-loop fiber optic gyroscope.

Appl Opt 2016 Jan;55(1):165-70

Analysis of the dead zone is among the intensive studies in a closed-loop fiber optic gyroscope. In a dead zone, a gyroscope cannot detect any rotation and produces a zero bias. In this study, an analysis of dead zone sources is performed in simulation and experiments. In general, the problem is mainly due to electrical cross coupling and phase modulation drift. Electrical cross coupling is caused by interference between modulation voltage and the photodetector. The cross-coupled signal produces spurious gyro bias and leads to a dead zone if it is larger than the input rate. Phase modulation drift as another dead zone source is due to the electrode contamination, the piezoelectric effect of the LiNbO3 substrate, or to organic fouling. This modulation drift lasts for a short or long period of time like a lead-lag filter response and produces gyro bias error, noise spikes, or dead zone. For a more detailed analysis, the cross-coupling effect and modulation phase drift are modeled as a filter and are simulated in both the open-loop and closed-loop modes. The sources of dead zone are more clearly analyzed in the simulation and experimental results.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1364/AO.55.000165DOI Listing
January 2016

Packet loss rate prediction using the sparse basis prediction model.

IEEE Trans Neural Netw 2007 May;18(3):950-4

The quality of multimedia communicated through the Internet is highly sensitive to packet loss. In this letter, we develop a time-series prediction model for the end-to-end packet loss rate (PLR). The estimate of the PLR is needed in several transmission control mechanisms such as the TCP-friendly congestion control mechanism for UDP traffic. In addition, it is needed to estimate the amount of redundancy for the forward error correction (FEC) mechanism. An accurate prediction would therefore be very valuable. We used a relatively novel prediction model called sparse basis prediction model. It is an adaptive nonlinear prediction approach, whereby a very large dictionary of possible inputs are extracted from the time series (for example, through moving averages, some nonlinear transformations, etc.). Only few of the very best inputs among the dictionary are selected and are combined linearly. An algorithm adaptively updates the input selection (as well as updates the weights) each time a new time sample arrives in a computationally efficient way. Simulation experiments indicate significantly better prediction performance for the sparse basis approach, as compared to other traditional nonlinear approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNN.2007.891681DOI Listing
May 2007