LMTRDA: Using logistic model tree to predict MiRNA-disease associations by fusing multi-source information of sequences and similarities

• DOI: 10.1371/journal.pcbi.1006865
• Publication Date: 2019-03-27T20:52:10Z
• Authors: Lei Wang; Zhu-Hong You; Xing Chen; Yang-Ming Li; Ya-Nan Dong; Li-Ping Li; Kai Zheng
• Keywords: Similarity (geometry); Decision tree model; Tree (set theory)

Emerging evidence has shown microRNAs (miRNAs) play an important role in human disease research. Identifying potential association among them is significant for the development of pathology, diagnose and therapy. However, only a tiny portion all miRNA-disease pairs current datasets are experimentally validated. This prompts high-precision computational methods to predict real interaction pairs. In this paper, we propose new model Logistic Model Tree predicting miRNA-Disease Association (LMTRDA) by fusing multi-source information including miRNA sequences, functional similarity, semantic known associations. particular, introduce sequence extract its features using natural language processing technique first time prediction model. cross-validation experiment, LMTRDA obtained 90.51% accuracy with 92.55% sensitivity at AUC 90.54% on HMDD V3.0 dataset. To further evaluate performance LMTRDA, compared it different classifier feature descriptor models. addition, also validate predictive ability diseases Breast Neoplasms, Neoplasms Lymphoma. As result, 28, 27 26 out top 30 miRNAs associated these were verified experiments kinds case studies. These experimental results demonstrate that reliable diseases.