AUTHOR=Qu Jia , Song Zihao , Cheng Xiaolong , Jiang Zhibin , Zhou Jie 

TITLE=A new integrated framework for the identification of potential virus–drug associations

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2023.1179414

DOI=10.3389/fmicb.2023.1179414

ISSN=1664-302X

ABSTRACT=With 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. Computa-tional models have ability to quickly predict potential reusable drug candidates to treat diseases. In this paper, two matrix decomposition-based methods, i.e., Matrix Decomposition with Hetero-geneous 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 five-fold cross validation were implemented to evaluate the performance of the proposed model based on datasets of DrugVirus that consist of 933 known associations be-tween 175 drugs and 95 viruses. The results showed that area under the receiver operating char-acteristics curve (AUC) of global LOOCV and local LOOCV are 0.9035 and 0.8786, respective-ly. The average AUC and the standard deviation of the five-fold cross validation for DrugVirus datasets is 0.8856+/-0.0032. We further implemented cross validate based on MDAD and aBio-film respectively to evaluate the performance of the model. In particle. MDAD (aBiofilm) da-taset contains 2470 (2884) known associations between 1373 (1470) drugs and 173 (140) mi-crobes. Also, two types of case studies were further carried out to verify the effectiveness of the model based on DrugVirus dataset and MDAD dataset. The results of case studies supported the effectiveness of MHBVDA in identifying potential virus-drug associations as well as predicting potential drug for new microbes.