Drug—target interaction prediction with a deep-learning-based model

Xie, Lingwei; Zhang, Zhongnan; He, Song; Bo, Xiaochen; Song, Xinyu

doi:10.1109/bibm.2017.8217693

Cited by 12 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Drug repositioning aims to discover new therapeutic uses for existing drugs, which takes advantage of drugs that have passed a significant number of toxicity and other tests. A variety of computational methods have been proposed for drug repositioning, which is much more efficient and effective, including machine-learning-based methods [6, 9], matrix decomposition-based methods [10, 13] and network-based methods [15, 16]. Most of the studies investigate pairwise associations among drugs, targets and diseases.…”

Section: Discussionmentioning

confidence: 99%

“…Then the machine learning models is trained on the feature vectors and further provides new predictions of associations [3, 4]. On account of the strong predictive power of deep learning methods in recent years, various deep learning models are applied to drug repositioning, including multi-layer perceptron [5, 6], deep belief network [7] stacked auto-encoder [8, 9], etc. When training the machine learning models, both positive and negative training samples should be provided.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting associations among drugs, targets and diseases by tensor decomposition for drug repositioning

Wang

Cheng

et al. 2019

BMC Bioinformatics

View full text Add to dashboard Cite

BackgroundDevelopment of new drugs is a time-consuming and costly process, and the cost is still increasing in recent years. However, the number of drugs approved by FDA every year per dollar spent on development is declining. Drug repositioning, which aims to find new use of existing drugs, attracts attention of pharmaceutical researchers due to its high efficiency. A variety of computational methods for drug repositioning have been proposed based on machine learning approaches, network-based approaches, matrix decomposition approaches, etc.ResultsWe propose a novel computational method for drug repositioning. We construct and decompose three-dimensional tensors, which consist of the associations among drugs, targets and diseases, to derive latent factors reflecting the functional patterns of the three kinds of entities. The proposed method outperforms several baseline methods in recovering missing associations. Most of the top predictions are validated by literature search and computational docking. Latent factors are used to cluster the drugs, targets and diseases into functional groups. Topological Data Analysis (TDA) is applied to investigate the properties of the clusters. We find that the latent factors are able to capture the functional patterns and underlying molecular mechanisms of drugs, targets and diseases. In addition, we focus on repurposing drugs for cancer and discover not only new therapeutic use but also adverse effects of the drugs. In the in-depth study of associations among the clusters of drugs, targets and cancer subtypes, we find there exist strong associations between particular clusters.ConclusionsThe proposed method is able to recover missing associations, discover new predictions and uncover functional clusters of drugs, targets and diseases. The clustering of drugs, targets and diseases, as well as the associations among the clusters, provides a new guiding framework for drug repositioning.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Predicting associations among drugs, targets and diseases by tensor decomposition for drug repositioning

Wang

Cheng

et al. 2019

BMC Bioinformatics

View full text Add to dashboard Cite

show abstract

“…The machine learning-based techniques may be loosely categorized as classification, matrix factorization, kernel methods, and network inference techniques. Support Vector Machine (SVM) is a classification method that has been used by (Manoochehri et al 2019;Lin et al 2019;Jung et al, 2020) Kernel methods mainly include the drug-target Kernel Method (PKM), network Laplacian regularized least squares method (NetLapRLS), and Regularized Least Squares with Kromecker Product Kernel (RLS-Kron) (Ye et al, 2020;Ezzat et al, 2016;Xie et al, 2017;Yasuo et al, 2018) established a bipartite local model and learned the drug-target interaction network, a typical network inference method. However, none of these basic methods can predict new drugs or targets (Wang et al, 2021;Xu et al, 2020) address this problem by interacting with neighbor information to expect new medicines or marks (Chen et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

“…Then, based on the grouped drug set, new hypotheses are put forth and the theoretical model of grouped Bayesian ranking is constructed. Finally, to improve the prediction of novel medications and targets, the article also includes neighbor information (Singh et al, 2022;2023;Chen et al, 2022;Zhu et al, 2020;Adam et al, 2020, Ezzat et al, 2016Xie et al, 2017;Xu et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

A Drug-Target Interaction Prediction Based on Supervised Probabilistic Classification

Singh,

Tiwari,

Swapna

et al. 2023

Journal of Computer Science

View full text Add to dashboard Cite

Bayesian ranking-based drug-target relationship prediction has achieved good results, but it ignores the relationship between drugs of the same target. A new method is proposed for drug-target relationship prediction based on groups by Appling Bayesian. According to the reality that drugs interacting with a specific target have similarities, a grouping strategy was introduced to make these similar drugs interact. A theoretical model based on the grouping strategy is derived in this study. The method is compared with five typical methods on five publicly available datasets and produces superior results to the compared methods. The impact of grouping interaction on the Bayesian ranking approach is examined in this study to create a grouped medication set; comparable pharmaceuticals that interact with the same target are first grouped based on this reality. Then, based on the grouped drug set, new hypotheses were put forth and the conceptual approach of grouped Bayesian ranking was constructed. Finally, to predict novel medications and targets, the article also includes neighbor information. The associated studies demonstrate that the strategy presented in this study outperforms the conventional performance techniques. Plans for further performance improvement through the creation of new comparable grouping objectives are included in future work.

show abstract

DTI-RCNN: New Efficient Hybrid Neural Network Model to Predict Drug–Target Interactions

Zheng

Song

et al. 2018

Artificial Neural Networks and Machine Learning – ICANN 2018

View full text Add to dashboard Cite

Drug—target interaction prediction with a deep-learning-based model

Cited by 12 publications

References 15 publications

Predicting associations among drugs, targets and diseases by tensor decomposition for drug repositioning

Predicting associations among drugs, targets and diseases by tensor decomposition for drug repositioning

A Drug-Target Interaction Prediction Based on Supervised Probabilistic Classification

DTI-RCNN: New Efficient Hybrid Neural Network Model to Predict Drug–Target Interactions

Contact Info

Product

Resources

About