Predicting Disease Related microRNA Based on Similarity and Topology

It is known that many diseases are caused by mutations or abnormalities in microRNA (miRNA). The usual method to predict miRNA disease relationships is to build a high-quality similarity network of diseases and miRNAs. All unobserved associations are ranked by their similarity scores, such that a hi...

Full description

Bibliographic Details
Main Authors: Zhihua Chen, Xinke Wang, Peng Gao, Hongju Liu, Bosheng Song
Format: Article
Language:English
Published: MDPI AG 2019-11-01
Series:Cells
Subjects:
Online Access:https://www.mdpi.com/2073-4409/8/11/1405
id doaj-af4a9aba43aa43a2bea17014c2fe8ef3
record_format Article
spelling doaj-af4a9aba43aa43a2bea17014c2fe8ef32020-11-24T22:10:06ZengMDPI AGCells2073-44092019-11-01811140510.3390/cells8111405cells8111405Predicting Disease Related microRNA Based on Similarity and TopologyZhihua Chen0Xinke Wang1Peng Gao2Hongju Liu3Bosheng Song4Institute of Computing Science and Technology, Guangzhou University, Guangzhou 510006, ChinaSchool of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, ChinaCollege of Information Technology and Computer Science, University of the Cordilleras, Baguio 2600, PhilippinesSchool of Information Science and Engineering, Hunan University, Changsha 410082, ChinaIt is known that many diseases are caused by mutations or abnormalities in microRNA (miRNA). The usual method to predict miRNA disease relationships is to build a high-quality similarity network of diseases and miRNAs. All unobserved associations are ranked by their similarity scores, such that a higher score indicates a greater probability of a potential connection. However, this approach does not utilize information within the network. Therefore, in this study, we propose a machine learning method, called STIM, which uses network topology information to predict disease−miRNA associations. In contrast to the conventional approach, STIM constructs features according to information on similarity and topology in networks and then uses a machine learning model to predict potential associations. To verify the reliability and accuracy of our method, we compared STIM to other classical algorithms. The results of fivefold cross validation demonstrated that STIM outperforms many existing methods, particularly in terms of the area under the curve. In addition, the top 30 candidate miRNAs recommended by STIM in a case study of lung neoplasm have been confirmed in previous experiments, which proved the validity of the method.https://www.mdpi.com/2073-4409/8/11/1405mirnanetwork embeddingheterogeneous networklink predictiontopology informationmachine learning
collection DOAJ
language English
format Article
sources DOAJ
author Zhihua Chen
Xinke Wang
Peng Gao
Hongju Liu
Bosheng Song
spellingShingle Zhihua Chen
Xinke Wang
Peng Gao
Hongju Liu
Bosheng Song
Predicting Disease Related microRNA Based on Similarity and Topology
Cells
mirna
network embedding
heterogeneous network
link prediction
topology information
machine learning
author_facet Zhihua Chen
Xinke Wang
Peng Gao
Hongju Liu
Bosheng Song
author_sort Zhihua Chen
title Predicting Disease Related microRNA Based on Similarity and Topology
title_short Predicting Disease Related microRNA Based on Similarity and Topology
title_full Predicting Disease Related microRNA Based on Similarity and Topology
title_fullStr Predicting Disease Related microRNA Based on Similarity and Topology
title_full_unstemmed Predicting Disease Related microRNA Based on Similarity and Topology
title_sort predicting disease related microrna based on similarity and topology
publisher MDPI AG
series Cells
issn 2073-4409
publishDate 2019-11-01
description It is known that many diseases are caused by mutations or abnormalities in microRNA (miRNA). The usual method to predict miRNA disease relationships is to build a high-quality similarity network of diseases and miRNAs. All unobserved associations are ranked by their similarity scores, such that a higher score indicates a greater probability of a potential connection. However, this approach does not utilize information within the network. Therefore, in this study, we propose a machine learning method, called STIM, which uses network topology information to predict disease−miRNA associations. In contrast to the conventional approach, STIM constructs features according to information on similarity and topology in networks and then uses a machine learning model to predict potential associations. To verify the reliability and accuracy of our method, we compared STIM to other classical algorithms. The results of fivefold cross validation demonstrated that STIM outperforms many existing methods, particularly in terms of the area under the curve. In addition, the top 30 candidate miRNAs recommended by STIM in a case study of lung neoplasm have been confirmed in previous experiments, which proved the validity of the method.
topic mirna
network embedding
heterogeneous network
link prediction
topology information
machine learning
url https://www.mdpi.com/2073-4409/8/11/1405
work_keys_str_mv AT zhihuachen predictingdiseaserelatedmicrornabasedonsimilarityandtopology
AT xinkewang predictingdiseaserelatedmicrornabasedonsimilarityandtopology
AT penggao predictingdiseaserelatedmicrornabasedonsimilarityandtopology
AT hongjuliu predictingdiseaserelatedmicrornabasedonsimilarityandtopology
AT boshengsong predictingdiseaserelatedmicrornabasedonsimilarityandtopology
_version_ 1725809266490605568