DeepFinder: An integration of feature-based and deep learning approach for DNA motif discovery

Lee, Nung Kion; Azizan, Farah Liyana; Wong, Yu Shiong; Omar, Norshafarina

doi:10.1080/13102818.2018.1438209

Cited by 13 publications

(6 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…De novo DNA motif discovery is an effective bioinformatic method for studying transcriptional gene regulation [11] and a number of motif discovery methods and tools currently exist. These include expectation-maximization methods, such as MEME [12] and Improbizer [13]; Gibbs sampling methods, such as BioProspector [14] and MotifSampler [15]; k-mer enumeration methods such as Weeder [16], DME [17], and DECOD [18]; ensemble methods such as W-ChIPMotifs [19], and GimmeMotifs [20]; and deep learning methods such as DanQ [21] and DeepFinder [22]. Based on the input types, motif discovery approaches can also be classified as either generative or discriminative.…”

Section: Introductionmentioning

confidence: 99%

Prediction pipeline for discovery of regulatory motifs associated with Brugia malayi molting

Grote

Liu

et al. 2020

PLoS Negl Trop Dis

View full text Add to dashboard Cite

Filarial nematodes can cause debilitating diseases in humans. They have complicated life cycles involving an insect vector and mammalian hosts, and they go through a number of developmental molts. While whole genome sequences of parasitic worms are now available, very little is known about transcription factor (TF) binding sites and their cognate transcription factors that play a role in regulating development. To address this gap, we developed a novel motif prediction pipeline, Emotif Alpha, that integrates ten different motif discovery algorithms, multiple statistical tests, and a comparative analysis of conserved elements between the filarial worms Brugia malayi and Onchocerca volvulus, and the free-living nematode Caenorhabditis elegans. We identified stage-specific TF binding motifs in B. malayi, with a particular focus on those potentially involved in the L3-L4 molt, a stage important for the establishment of infection in the mammalian host. Using an in vitro molting system, we tested and validated three of these motifs demonstrating the accuracy of the motif prediction pipeline.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction pipeline for discovery of regulatory motifs associated with Brugia malayi molting

Grote

Liu

et al. 2020

PLoS Negl Trop Dis

View full text Add to dashboard Cite

show abstract

“…Moreover, they cannot be applied to the entire genome efficiently. With the development of gene sequence project and gene expression profile, much efforts turn to recognize the TFBSs by computing-based methods [8][9][10].…”

Section: Introductionmentioning

confidence: 99%

Discovery of DNA Motif Utilising an Integrated Strategy Based on Random Projection and Particle Swarm Optimization

Sun

et al. 2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

During the process of gene expression and regulation, the DNA genetic information can be transferred to protein by means of transcription. The recognition of transcription factor binding sites can help to understand the evolutionary relations among different sequences. Thus, the problem of recognition of transcription factor binding sites, i.e., motif recognition, plays an important role for understanding the biological functions or meanings of sequences. However, when the established search space processes much noise subsequences, many optimization algorithms tend to be trapped into local optimum. In order to solve this problem, a particle swarm optimization and random projection-based algorithm (PSORPS) is proposed for recognizing DNA motifs. First, a random projection strategy is employed to filter the noise subsequences for constructing the objective space. Moreover, the sequence segments distributed in the majority of DNA sequences can be obtained and used for the population initialization of PSO. Then, the motifs of DNA sequences can be automatically searched by using a designed PSO algorithm in the constructed l-mer objective space. Finally, to alleviate the base deviation and further improve the recognition accuracy, the two operators of associated drift and independent drift are performed on the optimization results obtained by PSO. The experiments are conducted on real-world biological datasets, and the experimental results verify the effectiveness of the proposed algorithm.

show abstract

“…Deep learning is widely applied in bioinformatics area. For example, Lee et al [2] employed deep learning neural networks with features associated with binding sites to construct a DNA motif model. In addition, Khan et al [3] developed a method of classifying cancers to specific diagnostic categories based on their gene expression signatures using artificial neural networks (ANNs).…”

Section: Introductionmentioning

confidence: 99%

Integrative Deep Learning for Identifying Differentially Expressed (DE) Biomarkers

Lim

Bang

Kim

et al. 2019

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

As a large amount of genetic data are accumulated, an effective analytical method and a significant interpretation are required. Recently, various methods of machine learning have emerged to process genetic data. In addition, machine learning analysis tools using statistical models have been proposed. In this study, we propose adding an integrated layer to the deep learning structure, which would enable the effective analysis of genetic data and the discovery of significant biomarkers of diseases. We conducted a simulation study in order to compare the proposed method with metalogistic regression and meta-SVM methods. The objective function with lasso penalty is used for parameter estimation, and the Youden J index is used for model comparison. The simulation results indicate that the proposed method is more robust for the variance of the data than metalogistic regression and meta-SVM methods. We also conducted real data (breast cancer data (TCGA)) analysis. Based on the results of gene set enrichment analysis, we obtained that TCGA multiple omics data involve significantly enriched pathways which contain information related to breast cancer. Therefore, it is expected that the proposed method will be helpful to discover biomarkers.

show abstract

DeepFinder: An integration of feature-based and deep learning approach for DNA motif discovery

Cited by 13 publications

References 52 publications

Prediction pipeline for discovery of regulatory motifs associated with Brugia malayi molting

Prediction pipeline for discovery of regulatory motifs associated with Brugia malayi molting

Discovery of DNA Motif Utilising an Integrated Strategy Based on Random Projection and Particle Swarm Optimization

Integrative Deep Learning for Identifying Differentially Expressed (DE) Biomarkers

Contact Info

Product

Resources

About