scholarly journals 6mAPred-MSFF: A Deep Learning Model for Predicting DNA N6-Methyladenine Sites across Species Based on a Multi-Scale Feature Fusion Mechanism

2021 ◽  
Vol 11 (16) ◽  
pp. 7731
Author(s):  
Rao Zeng ◽  
Minghong Liao
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Prediction Accuracy ◽  
Cross Validation ◽  
Feature Fusion ◽  
Experimental Comparison ◽  
Scale Feature ◽  
Learning Framework ◽  
Multi Scale ◽  
Fold Cross Validation

DNA methylation is one of the most extensive epigenetic modifications. DNA N6-methyladenine (6mA) plays a key role in many biology regulation processes. An accurate and reliable genome-wide identification of 6mA sites is crucial for systematically understanding its biological functions. Some machine learning tools can identify 6mA sites, but their limited prediction accuracy and lack of robustness limit their usability in epigenetic studies, which implies the great need of developing new computational methods for this problem. In this paper, we developed a novel computational predictor, namely the 6mAPred-MSFF, which is a deep learning framework based on a multi-scale feature fusion mechanism to identify 6mA sites across different species. In the predictor, we integrate the inverted residual block and multi-scale attention mechanism to build lightweight and deep neural networks. As compared to existing predictors using traditional machine learning, our deep learning framework needs no prior knowledge of 6mA or manually crafted sequence features and sufficiently capture better characteristics of 6mA sites. By benchmarking comparison, our deep learning method outperforms the state-of-the-art methods on the 5-fold cross-validation test on the seven datasets of six species, demonstrating that the proposed 6mAPred-MSFF is more effective and generic. Specifically, our proposed 6mAPred-MSFF gives the sensitivity and specificity of the 5-fold cross-validation on the 6mA-rice-Lv dataset as 97.88% and 94.64%, respectively. Our model trained with the rice data predicts well the 6mA sites of other five species: Arabidopsis thaliana, Fragaria vesca, Rosa chinensis, Homo sapiens, and Drosophila melanogaster with a prediction accuracy 98.51%, 93.02%, and 91.53%, respectively. Moreover, via experimental comparison, we explored performance impact by training and testing our proposed model under different encoding schemes and feature descriptors.

scholarly journals Deep6mA: A deep learning framework for exploring similar patterns in DNA N6-methyladenine sites across different species

2021 ◽  
Vol 17 (2) ◽  
pp. e1008767
Author(s):  
Zutan Li ◽  
Hangjin Jiang ◽  
Lingpeng Kong ◽  
Yuanyuan Chen ◽  
Kun Lang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Prediction Accuracy ◽  
Cross Validation ◽  
Biological Processes ◽  
Biological Functions ◽  
Learning Framework ◽  
Wide Range ◽  
Genomic Scale ◽  
Downstream Gene Expression ◽  
Fold Cross Validation

N6-methyladenine (6mA) is an important DNA modification form associated with a wide range of biological processes. Identifying accurately 6mA sites on a genomic scale is crucial for under-standing of 6mA’s biological functions. However, the existing experimental techniques for detecting 6mA sites are cost-ineffective, which implies the great need of developing new computational methods for this problem. In this paper, we developed, without requiring any prior knowledge of 6mA and manually crafted sequence features, a deep learning framework named Deep6mA to identify DNA 6mA sites, and its performance is superior to other DNA 6mA prediction tools. Specifically, the 5-fold cross-validation on a benchmark dataset of rice gives the sensitivity and specificity of Deep6mA as 92.96% and 95.06%, respectively, and the overall prediction accuracy is 94%. Importantly, we find that the sequences with 6mA sites share similar patterns across different species. The model trained with rice data predicts well the 6mA sites of other three species: Arabidopsis thaliana, Fragaria vesca and Rosa chinensis with a prediction accuracy over 90%. In addition, we find that (1) 6mA tends to occur at GAGG motifs, which means the sequence near the 6mA site may be conservative; (2) 6mA is enriched in the TATA box of the promoter, which may be the main source of its regulating downstream gene expression.

scholarly journals Deep6mA: a deep learning framework for exploring similar patterns in DNA N6-methyladenine sites across different species

2019 ◽  
Author(s):  
Zutan Li ◽  
Hangjin Jiang ◽  
Lingpeng Kong ◽  
Yuanyuan Chen ◽  
Liangyun Zhang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Prediction Accuracy ◽  
Cross Validation ◽  
Biological Processes ◽  
Biological Functions ◽  
Learning Framework ◽  
Wide Range ◽  
Genomic Scale ◽  
Downstream Gene Expression ◽  
Fold Cross Validation

ABSTRACTN6-methyladenin(6mA) is an important DNA modification form associated with a wide range of biological processes. Identifying accurately 6mA sites on a genomic scale is crucial for understanding of 6mA’s biological functions. In this paper, we developed, without requiring any prior knowledge of 6mA and manually crafted sequence features, a deep learning framework named Deep6mA to identify DNA 6mA sites, and its performance is superior to other DNA 6mA prediction tools. Specifically, the 5-fold cross-validation on a benchmark dataset of rice gives the sensitivity and specificity of Deep6mA as 92.96% and 95.06%, respectively, and the overall prediction accuracy is 94%. Importantly, we find that the sequences with 6mA sites share similar patterns across different species. The model trained with rice data predicts well the 6mA sites of other three species: Arabidopsis thaliana, Fragaria vesca, and Rosa chinensis, with a prediction accuracy over 90%. In addition, we find that (1) 6mA tends to occur at GAGG motifs, which means the sequence near the 6mA site may be conservative; (2) 6mA is enriched in the TATA box of the promoter, which may be the main source of its regulating downstream gene expression.

MUFFIN: multi-scale feature fusion for drug–drug interaction prediction

2021 ◽  
Author(s):  
Yujie Chen ◽  
Tengfei Ma ◽  
Xixi Yang ◽  
Jianmin Wang ◽  
Bosheng Song ◽  
...  
Keyword(s):  
Molecular Structure ◽  
Deep Learning ◽  
Medical Information ◽  
Feature Fusion ◽  
Molecular Graph ◽  
Knowledge Graph ◽  
Sequence Information ◽  
Learning Models ◽  
Scale Feature ◽  
Multi Scale

Abstract Motivation Adverse drug–drug interactions (DDIs) are crucial for drug research and mainly cause morbidity and mortality. Thus, the identification of potential DDIs is essential for doctors, patients and the society. Existing traditional machine learning models rely heavily on handcraft features and lack generalization. Recently, the deep learning approaches that can automatically learn drug features from the molecular graph or drug-related network have improved the ability of computational models to predict unknown DDIs. However, previous works utilized large labeled data and merely considered the structure or sequence information of drugs without considering the relations or topological information between drug and other biomedical objects (e.g. gene, disease and pathway), or considered knowledge graph (KG) without considering the information from the drug molecular structure. Results Accordingly, to effectively explore the joint effect of drug molecular structure and semantic information of drugs in knowledge graph for DDI prediction, we propose a multi-scale feature fusion deep learning model named MUFFIN. MUFFIN can jointly learn the drug representation based on both the drug-self structure information and the KG with rich bio-medical information. In MUFFIN, we designed a bi-level cross strategy that includes cross- and scalar-level components to fuse multi-modal features well. MUFFIN can alleviate the restriction of limited labeled data on deep learning models by crossing the features learned from large-scale KG and drug molecular graph. We evaluated our approach on three datasets and three different tasks including binary-class, multi-class and multi-label DDI prediction tasks. The results showed that MUFFIN outperformed other state-of-the-art baselines. Availability and implementation The source code and data are available at https://github.com/xzenglab/MUFFIN.

scholarly journals lncRNA_Mdeep: An Alignment-Free Predictor for Distinguishing Long Non-Coding RNAs from Protein-Coding Transcripts by Multimodal Deep Learning

2020 ◽  
Vol 21 (15) ◽  
pp. 5222 ◽  
Author(s):  
Xiao-Nan Fan ◽  
Shao-Wu Zhang ◽  
Song-Yao Zhang ◽  
Jin-Jie Ni
Keyword(s):  
Deep Learning ◽  
Prediction Accuracy ◽  
Protein Coding ◽  
Learning Framework ◽  
Alignment Free ◽  
Independent Test ◽  
Non Coding Rnas ◽  
High Level ◽  
Fold Cross Validation ◽  
Human Complex

Long non-coding RNAs (lncRNAs) play crucial roles in diverse biological processes and human complex diseases. Distinguishing lncRNAs from protein-coding transcripts is a fundamental step for analyzing the lncRNA functional mechanism. However, the experimental identification of lncRNAs is expensive and time-consuming. In this study, we presented an alignment-free multimodal deep learning framework (namely lncRNA_Mdeep) to distinguish lncRNAs from protein-coding transcripts. LncRNA_Mdeep incorporated three different input modalities, then a multimodal deep learning framework was built for learning the high-level abstract representations and predicting the probability whether a transcript was lncRNA or not. LncRNA_Mdeep achieved 98.73% prediction accuracy in a 10-fold cross-validation test on humans. Compared with other eight state-of-the-art methods, lncRNA_Mdeep showed 93.12% prediction accuracy independent test on humans, which was 0.94%~15.41% higher than that of other eight methods. In addition, the results on 11 cross-species datasets showed that lncRNA_Mdeep was a powerful predictor for predicting lncRNAs.

scholarly journals lncRNA_Mdeep: an alignment-free predictor for long non-coding RNAs identification by multimodal deep learning

2020 ◽  
Author(s):  
Xiao-Nan Fan ◽  
Shao-Wu Zhang ◽  
Song-Yao Zhang ◽  
Jin-Jie Ni
Keyword(s):  
Deep Learning ◽  
Prediction Accuracy ◽  
Protein Coding ◽  
Learning Framework ◽  
Alignment Free ◽  
Independent Test ◽  
Non Coding Rnas ◽  
High Level ◽  
Fold Cross Validation ◽  
Human Complex

Abstract Background: Long non-coding RNAs (lncRNAs) play crucial roles in diverse biological processes and human complex diseases. Distinguishing lncRNAs from protein-coding transcripts is a fundamental step for analyzing lncRNA functional mechanism. However, the experimental identification of lncRNAs is expensive and time-consuming. Results: In this study, we present an alignment-free multimodal deep learning framework (namely lncRNA_Mdeep) to distinguish lncRNAs from protein-coding transcripts. LncRNA_Mdeep incorporates three different input modalities (i.e. OFH modality, k-mer modality, and sequence modality), then a multimodal deep learning framework is built for learning the high-level abstract representations and predicting the probability whether a transcript is lncRNA or not. Conclusions: LncRNA_Mdeep achieves 98.73% prediction accuracy in 10-fold cross-validation test on human. Compared with other eight state-of-the-art methods, lncRNA_Mdeep shows 93.12% prediction accuracy independent test on human, which is 0.94%~15.41% higher than that of other eight methods. In addition, the results on 11 cross-species datasets show that lncRNA_Mdeep is a powerful predictor for identifying lncRNAs. The source code can be downloaded from https://github.com/NWPU-903PR/lncRNA_Mdeep.

Prediction models establishment and comparison for guiding force of high-temperature superconducting maglev based on deep learning algorithms

2021 ◽  
Author(s):  
Zhihao Ke ◽  
Xiaoning Liu ◽  
Yining Chen ◽  
Hongfu Shi ◽  
Zigang Deng
Keyword(s):  
Deep Learning ◽  
High Temperature ◽  
Prediction Accuracy ◽  
Cross Validation ◽  
Learning Algorithms ◽  
Learning Rate ◽  
Force Prediction ◽  
Validation Method ◽  
Hts Maglev ◽  
Fold Cross Validation

Abstract By the merits of self-stability and low energy consumption, high temperature superconducting (HTS) maglev has the potential to become a novel type of transportation mode. As a key index to guarantee the lateral self-stability of HTS maglev, guiding force has strong non-linearity and is determined by multitudinous factors, and these complexities impede its further researches. Compared to traditional finite element and polynomial fitting method, the prosperity of deep learning algorithms could provide another guiding force prediction approach, but the verification of this approach is still blank. Therefore, this paper establishes 5 different neural network models (RBF, DNN, CNN, RNN, LSTM) to predict HTS maglev guiding force, and compares their prediction efficiency based on 3720 pieces of collected data. Meanwhile, two adaptively iterative algorithms for parameters matrix and learning rate adjustment are proposed, which could effectively reduce computing time and unnecessary iterations. And according to the results, it is revealed that, the DNN model shows the best fitting goodness, while the LSTM model displays the smoothest fitting curve on guiding force prediction. Based on this discovery, the effects of learning rate and iterations on prediction accuracy of the constructed DNN model are studied. And the learning rate and iterations at the highest guiding force prediction accuracy are 0.00025 and 90000, respectively. Moreover, the K-fold cross validation method is also applied to this DNN model, whose result manifests the generalization and robustness of this DNN model. The imperative of K-fold cross validation method to ensure universality of guiding force prediction model is likewise assessed. This paper firstly combines HTS maglev guiding force prediction with deep learning algorithms considering different field cooling height, real-time magnetic flux density, liquid nitrogen temperature and motion direction of bulk. Additionally, this paper gives a convenient and efficient method for HTS guiding force prediction and parameter optimization.

Sign in / Sign up

Export Citation Format

Share Document