{"title":"Efficient prediction of drug–drug interaction using deep learning models","authors":"Prashant Kumar Shukla, Piyush Kumar Shukla, Poonam Sharma, Paresh Rawat, Jashwant Samar, Rahul Moriwal, Manjit Kaur","doi":"10.1049/iet-syb.2019.0116","DOIUrl":null,"url":null,"abstract":"<div>\n <p>A drug–drug interaction or drug synergy is extensively utilised for cancer treatment. However, prediction of drug–drug interaction is defined as an ill-posed problem, because manual testing is only implementable on small group of drugs. Predicting the drug–drug interaction score has been a popular research topic recently. Recently many machine learning models have proposed in the literature to predict the drug–drug interaction score efficiently. However, these models suffer from the over-fitting issue. Therefore, these models are not so-effective for predicting the drug–drug interaction score. In this work, an integrated convolutional mixture density recurrent neural network is proposed and implemented. The proposed model integrates convolutional neural networks, recurrent neural networks and mixture density networks. Extensive comparative analysis reveals that the proposed model significantly outperforms the competitive models.</p>\n </div>","PeriodicalId":50379,"journal":{"name":"IET Systems Biology","volume":null,"pages":null},"PeriodicalIF":1.9000,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1049/iet-syb.2019.0116","citationCount":"70","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IET Systems Biology","FirstCategoryId":"99","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1049/iet-syb.2019.0116","RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"CELL BIOLOGY","Score":null,"Total":0}
引用次数: 70
Abstract
A drug–drug interaction or drug synergy is extensively utilised for cancer treatment. However, prediction of drug–drug interaction is defined as an ill-posed problem, because manual testing is only implementable on small group of drugs. Predicting the drug–drug interaction score has been a popular research topic recently. Recently many machine learning models have proposed in the literature to predict the drug–drug interaction score efficiently. However, these models suffer from the over-fitting issue. Therefore, these models are not so-effective for predicting the drug–drug interaction score. In this work, an integrated convolutional mixture density recurrent neural network is proposed and implemented. The proposed model integrates convolutional neural networks, recurrent neural networks and mixture density networks. Extensive comparative analysis reveals that the proposed model significantly outperforms the competitive models.
期刊介绍:
IET Systems Biology covers intra- and inter-cellular dynamics, using systems- and signal-oriented approaches. Papers that analyse genomic data in order to identify variables and basic relationships between them are considered if the results provide a basis for mathematical modelling and simulation of cellular dynamics. Manuscripts on molecular and cell biological studies are encouraged if the aim is a systems approach to dynamic interactions within and between cells.
The scope includes the following topics:
Genomics, transcriptomics, proteomics, metabolomics, cells, tissue and the physiome; molecular and cellular interaction, gene, cell and protein function; networks and pathways; metabolism and cell signalling; dynamics, regulation and control; systems, signals, and information; experimental data analysis; mathematical modelling, simulation and theoretical analysis; biological modelling, simulation, prediction and control; methodologies, databases, tools and algorithms for modelling and simulation; modelling, analysis and control of biological networks; synthetic biology and bioengineering based on systems biology.