Drug-Target Interaction Prediction via Dual Laplacian Graph Regularized Logistic Matrix Factorization

Wang, Aizhen; Wang, Minhui

doi:10.1155/2021/5599263

Cited by 5 publications

(4 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This phenomenon with AUROC is known to occur when the number of positive labels is much lesser than that of negative labels, and AUPR is more appropriate than AUROC. 17,37,38 Through these comparisons, NRLMF was found to perform better in CVS1 than in CVS2 and CVS3, which was consistent with the findings of the original study reviewed in the present study. 7 However, evaluation of CVS3 using the drug−protein benchmark dataset showed poor results for all methods.…”

Section: Resultssupporting

confidence: 90%

See 1 more Smart Citation

NRBdMF: A Recommendation Algorithm for Predicting Drug Effects Considering Directionality

Azuma

Mizuno

Kusuhara

2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Predicting the novel effects of drugs based on information about approved drugs can be regarded as a recommendation system. Matrix factorization is one of the most used recommendation systems, and various algorithms have been devised for it. A literature survey and summary of existing algorithms for predicting drug effects demonstrated that most such methods, including neighborhood regularized logistic matrix factorization, which was the best performer in benchmark tests, used a binary matrix that considers only the presence or absence of interactions. However, drug effects are known to have two opposite aspects, such as side effects and therapeutic effects. In the present study, we proposed using neighborhood regularized bidirectional matrix factorization (NRBdMF) to predict drug effects by incorporating bidirectionality, which is a characteristic property of drug effects. We used this proposed method for predicting side effects using a matrix that considered the bidirectionality of drug effects, in which known side effects were assigned a positive (+1) label and known treatment effects were assigned a negative (−1) label. The NRBdMF model, which utilizes drug bidirectional information, achieved enrichment of side effects at the top and indications at the bottom of the prediction list. This first attempt to consider the bidirectional nature of drug effects using NRBdMF showed that it reduced false positives and produced a highly interpretable output.

show abstract

Section: Resultssupporting

confidence: 90%

“…CVS1, CVS2, and CVS3 were settings for the prediction of new pairs, drugs, and targets, respectively. These three performance evaluation scenarios were found to be used widely. − This evaluation metric was the most used throughout the survey.…”

Section: Methodsmentioning

confidence: 99%

NRBdMF: A Recommendation Algorithm for Predicting Drug Effects Considering Directionality

Azuma

Mizuno

Kusuhara

2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…There are three branches of machine learning methods for predicting DTIs: similarity-based methods, deep learning methods, and feature selection methods. Similarity/distance-based methods mainly use inter-sample similarity or distance [20][21][22]. Yamanishi et al [23] developed a bipartite graph model to predict DTIs using a supervised approach to learn known drug-target relationships [24,25].…”

Section: Introductionmentioning

confidence: 99%

SSELM-neg: spherical search-based extreme learning machine for drug–target interaction prediction

Ren

et al. 2023

BMC Bioinformatics

View full text Add to dashboard Cite

Background The experimental verification of a drug discovery process is expensive and time-consuming. Therefore, efficiently and effectively identifying drug–target interactions (DTIs) has been the focus of research. At present, many machine learning algorithms are used for predicting DTIs. The key idea is to train the classifier using an existing DTI to predict a new or unknown DTI. However, there are various challenges, such as class imbalance and the parameter optimization of many classifiers, that need to be solved before an optimal DTI model is developed. Methods In this study, we propose a framework called SSELM-neg for DTI prediction, in which we use a screening approach to choose high-quality negative samples and a spherical search approach to optimize the parameters of the extreme learning machine. Results The results demonstrated that the proposed technique outperformed other state-of-the-art methods in 10-fold cross-validation experiments in terms of the area under the receiver operating characteristic curve (0.986, 0.993, 0.988, and 0.969) and AUPR (0.982, 0.991, 0.982, and 0.946) for the enzyme dataset, G-protein coupled receptor dataset, ion channel dataset, and nuclear receptor dataset, respectively. Conclusion The screening approach produced high-quality negative samples with the same number of positive samples, which solved the class imbalance problem. We optimized an extreme learning machine using a spherical search approach to identify DTIs. Therefore, our models performed better than other state-of-the-art methods.

show abstract

“…There are three branches of machine learning methods for predicting DTIs: similarity-based methods, deep learning methods and feature selection methods. Similarity/distance-based methods mainly utilize inter-sample similarity or distance [18,19,20] . Yamanishi et al [21] developed a bipartite graph model to predict drug-target interactions, using a supervised approach to learn known drugtarget relationships [22,23]; Buza et al proposed ECkNN/HLM , a KNN method (hub-aware regression technique) with error correction to mitigate the harmful effects of bad hubs [24,25,26]; Mei et al proposed BLM-NII, an inference integrated into a BLM approach, to deal with a new candidate problem for pure BLM [27].…”

Section: Introductionmentioning

confidence: 99%

SSELM-neg: Spherical search-based extreme learning machine for drug-target interaction prediction

Ren

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Experimental verification of a drug discovery process is expensive and time-consuming. Therefore, efficiently and effectively identify drug–target interactions (DTIs) has been research focus. In present, many machine learning algorithms are used for predicting drug-target interactions. The key idea is to train the classifier using existing DTI so as to predict new or unknown DTI. However, there are various challenges such as class imbalance and parameter optimization of many classifiers that need to be solved before developing an optimal drug-target interaction model. Methods: In this paper, we propose a framework named SSELM-neg for drug-target interaction prediction, in which a screening approach was used to choose high quality negative samples and a spherical search approach is used to optimize the parameters of the extreme learning machine. Results: Results show that the performance of the proposed technique outperforms the other state-of-the-art methods in 10-fold cross-validation experiments in terms of AUC(0.986, 0.993, 0.988, 0.969 ) and AUPR(0.982, 0.991, 0.982, 0.946) for E dataset, GPCR dataset, IC dataset and NR dataset. Conclusion: Screening approach produce the high quality negative samples with the same numbers of the positive samples, which solve the class imbalance problem. Extreme learning machine was optimized by a spherical search approach in identifying DTIs. Therefore, our models achieve better performance than other state-of-the-art methods.

show abstract

Drug-Target Interaction Prediction via Dual Laplacian Graph Regularized Logistic Matrix Factorization

Cited by 5 publications

References 53 publications

NRBdMF: A Recommendation Algorithm for Predicting Drug Effects Considering Directionality

NRBdMF: A Recommendation Algorithm for Predicting Drug Effects Considering Directionality

SSELM-neg: spherical search-based extreme learning machine for drug–target interaction prediction

SSELM-neg: Spherical search-based extreme learning machine for drug-target interaction prediction

Contact Info

Product

Resources

About