AUTHOR=Wu Qingwen , Wang Yutian , Gao Zhen , Ni Jiancheng , Zheng Chunhou 

TITLE=MSCHLMDA: Multi-Similarity Based Combinative Hypergraph Learning for Predicting MiRNA-Disease Association

JOURNAL=Frontiers in Genetics

VOLUME=Volume 11 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2020.00354

DOI=10.3389/fgene.2020.00354

ISSN=1664-8021

ABSTRACT=Accumulating biological and clinical reports have confirmed the important associations between microRNAs (miRNAs) and a lot of complex human diseases. Predicting disease-related miRNAs is beneficial to understand the molecular mechanisms of human diseases at the  miRNA level,  and promote finding biomarkers for human disease diagnosis, treatment, and prevention. However, the challenge for researchers is how to effectively combine different datasets and make credible prediction. In this work, we proposed the method of Multi-Similarity based Combinative Hypergraph Learning for Predicting MiRNA-disease Association(MSCHLMDA). In particular, we firstly extracted complex features by two measures for each miRNA-disease pair. Then, K-Nearest Neighbor (KNN) and K-means algorithm were used to construct two different hypergraphs. Finally, combinative hypergraph learning results were utilized for predicting miRNA–disease association. In order to evaluate the prediction effect of our method, leave-one-out cross validation and 5-fold cross validation were implemented, and the experimental results showed that our method had better performance than former typical methods. Moreover, three kinds of case studies on different human complex diseases were also carried out, which further demonstrated the predictive power of MSCHLMDA . It is anticipated that MSCHLMDA would become an excellent complement to the biomedical research field in the future.