Abstract
Increasing evidences have indicated that microRNAs (miRNAs) are functionally associated with the development and progression of various complex human diseases. However, the roles of miRNAs in multiple biological processes or various diseases and their underlying molecular mechanisms still have not been fully understood yet. Predicting potential miRNA-disease associations by integrating various heterogeneous biological datasets is of great significance to the biomedical research. Computational methods could obtain potential miRNA-disease associations in a short time, which significantly reduce the experimental time and cost. Considering the limitations in previous computational methods, we developed the model of Within and Between Score for MiRNA-Disease Association prediction (WBSMDA) to predict potential miRNAs associated with various complex diseases. WBSMDA could be applied to the diseases without any known related miRNAs. The AUC of 0.8031 based on Leave-one-out cross validation has demonstrated its reliable performance. WBSMDA was further applied to Colon Neoplasms, Prostate Neoplasms, and Lymphoma for the identification of their potential related miRNAs. As a result, 90%, 84%, and 80% of predicted miRNA-disease pairs in the top 50 prediction list for these three diseases have been confirmed by recent experimental literatures, respectively. It is anticipated that WBSMDA would be a useful resource for potential miRNA-disease association identification.
Highlights
Accumulating studies have shown that miRNAs play a critical role in many important biological processes, including cell proliferation[9], development[10], differentiation[11], and apoptosis[12], metabolism[13,14], aging[13,14], signal transduction[15], viral infection[11] and so on
Based on the assumption that functional similar miRNAs tend to interact with similar diseases, we developed the model of Within and Between Score for MiRNA-Disease Association prediction (WBSMDA) by integrating known miRNA-disease associations, miRNA functional similarity network, disease semantic similarity network, and Gaussian interaction profile kernel similarity network to uncover the potential disease-miRNA associations
LOOCV was implemented on known miRNA-disease associations obtained from HMDD51 to evaluate the predictive performance of WBSMDA
Summary
LOOCV was implemented on known miRNA-disease associations obtained from HMDD51 to evaluate the predictive performance of WBSMDA. Taking lymphomas as a case study to implement WBSMDA for potential miRNA-disease association prediction, top ten potential lymphoma-associated miRNAs in the prediction list were all successfully verified based on recent experimental reports (See Table 1 and Supplementary Table 2). Case studies of Colon Neoplasms, lymphoma and Prostate Neoplasms were implemented and 90%, 84%, and 80% of predicted miRNA-disease pairs in the top 50 prediction list for these three important diseases have been confirmed based on recent experimental literatures, respectively. The prediction performance of WBSMDA will be further improved by integrating more reliable biological datasets and obtaining more known miRNA-disease associations. How to more reasonably integrate similarity measure and integrate Within-Score and Between-Score to calculate the association score of miRNA-disease pair deserve further research in the future
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
