A new integrated framework for the identification of potential virus–drug associations
IntroductionWith the increasingly serious problem of antiviral drug resistance, drug repurposing offers a time-efficient and cost-effective way to find potential therapeutic agents for disease. Computational models have the ability to quickly predict potential reusable drug candidates to treat disea...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2023-08-01
|
Series: | Frontiers in Microbiology |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fmicb.2023.1179414/full |
_version_ | 1797739369515188224 |
---|---|
author | Jia Qu Zihao Song Xiaolong Cheng Zhibin Jiang Jie Zhou |
author_facet | Jia Qu Zihao Song Xiaolong Cheng Zhibin Jiang Jie Zhou |
author_sort | Jia Qu |
collection | DOAJ |
description | IntroductionWith the increasingly serious problem of antiviral drug resistance, drug repurposing offers a time-efficient and cost-effective way to find potential therapeutic agents for disease. Computational models have the ability to quickly predict potential reusable drug candidates to treat diseases.MethodsIn this study, two matrix decomposition-based methods, i.e., Matrix Decomposition with Heterogeneous Graph Inference (MDHGI) and Bounded Nuclear Norm Regularization (BNNR), were integrated to predict anti-viral drugs. Moreover, global leave-one-out cross-validation (LOOCV), local LOOCV, and 5-fold cross-validation were implemented to evaluate the performance of the proposed model based on datasets of DrugVirus that consist of 933 known associations between 175 drugs and 95 viruses.ResultsThe results showed that the area under the receiver operating characteristics curve (AUC) of global LOOCV and local LOOCV are 0.9035 and 0.8786, respectively. The average AUC and the standard deviation of the 5-fold cross-validation for DrugVirus datasets are 0.8856 ± 0.0032. We further implemented cross-validation based on MDAD and aBiofilm, respectively, to evaluate the performance of the model. In particle, MDAD (aBiofilm) dataset contains 2,470 (2,884) known associations between 1,373 (1,470) drugs and 173 (140) microbes. In addition, two types of case studies were carried out further to verify the effectiveness of the model based on the DrugVirus and MDAD datasets. The results of the case studies supported the effectiveness of MHBVDA in identifying potential virus-drug associations as well as predicting potential drugs for new microbes. |
first_indexed | 2024-03-12T13:57:02Z |
format | Article |
id | doaj.art-a0121a68e82d4c349e7a1ccb19968931 |
institution | Directory Open Access Journal |
issn | 1664-302X |
language | English |
last_indexed | 2024-03-12T13:57:02Z |
publishDate | 2023-08-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Microbiology |
spelling | doaj.art-a0121a68e82d4c349e7a1ccb199689312023-08-22T13:53:25ZengFrontiers Media S.A.Frontiers in Microbiology1664-302X2023-08-011410.3389/fmicb.2023.11794141179414A new integrated framework for the identification of potential virus–drug associationsJia Qu0Zihao Song1Xiaolong Cheng2Zhibin Jiang3Jie Zhou4School of Computer Science and Artificial Intelligence, Changzhou University, Changzhou, Jiangsu, ChinaSchool of Computer Science and Artificial Intelligence, Changzhou University, Changzhou, Jiangsu, ChinaSchool of Computer Science and Artificial Intelligence, Changzhou University, Changzhou, Jiangsu, ChinaSchool of Computer Science and Engineering, Shaoxing University, Shaoxing, Zhejiang, ChinaSchool of Computer Science and Engineering, Shaoxing University, Shaoxing, Zhejiang, ChinaIntroductionWith the increasingly serious problem of antiviral drug resistance, drug repurposing offers a time-efficient and cost-effective way to find potential therapeutic agents for disease. Computational models have the ability to quickly predict potential reusable drug candidates to treat diseases.MethodsIn this study, two matrix decomposition-based methods, i.e., Matrix Decomposition with Heterogeneous Graph Inference (MDHGI) and Bounded Nuclear Norm Regularization (BNNR), were integrated to predict anti-viral drugs. Moreover, global leave-one-out cross-validation (LOOCV), local LOOCV, and 5-fold cross-validation were implemented to evaluate the performance of the proposed model based on datasets of DrugVirus that consist of 933 known associations between 175 drugs and 95 viruses.ResultsThe results showed that the area under the receiver operating characteristics curve (AUC) of global LOOCV and local LOOCV are 0.9035 and 0.8786, respectively. The average AUC and the standard deviation of the 5-fold cross-validation for DrugVirus datasets are 0.8856 ± 0.0032. We further implemented cross-validation based on MDAD and aBiofilm, respectively, to evaluate the performance of the model. In particle, MDAD (aBiofilm) dataset contains 2,470 (2,884) known associations between 1,373 (1,470) drugs and 173 (140) microbes. In addition, two types of case studies were carried out further to verify the effectiveness of the model based on the DrugVirus and MDAD datasets. The results of the case studies supported the effectiveness of MHBVDA in identifying potential virus-drug associations as well as predicting potential drugs for new microbes.https://www.frontiersin.org/articles/10.3389/fmicb.2023.1179414/fulldrugvirusassociation predictionmatrix decompositionBounded Nuclear Norm Regularizationensemble learning |
spellingShingle | Jia Qu Zihao Song Xiaolong Cheng Zhibin Jiang Jie Zhou A new integrated framework for the identification of potential virus–drug associations Frontiers in Microbiology drug virus association prediction matrix decomposition Bounded Nuclear Norm Regularization ensemble learning |
title | A new integrated framework for the identification of potential virus–drug associations |
title_full | A new integrated framework for the identification of potential virus–drug associations |
title_fullStr | A new integrated framework for the identification of potential virus–drug associations |
title_full_unstemmed | A new integrated framework for the identification of potential virus–drug associations |
title_short | A new integrated framework for the identification of potential virus–drug associations |
title_sort | new integrated framework for the identification of potential virus drug associations |
topic | drug virus association prediction matrix decomposition Bounded Nuclear Norm Regularization ensemble learning |
url | https://www.frontiersin.org/articles/10.3389/fmicb.2023.1179414/full |
work_keys_str_mv | AT jiaqu anewintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT zihaosong anewintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT xiaolongcheng anewintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT zhibinjiang anewintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT jiezhou anewintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT jiaqu newintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT zihaosong newintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT xiaolongcheng newintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT zhibinjiang newintegratedframeworkfortheidentificationofpotentialvirusdrugassociations AT jiezhou newintegratedframeworkfortheidentificationofpotentialvirusdrugassociations |