AUTHOR=Wen Xiao , Gao Lin , Hu Yuxuan 

TITLE=LAceModule: Identification of Competing Endogenous RNA Modules by Integrating Dynamic Correlation

JOURNAL=Frontiers in Genetics

VOLUME=Volume 11 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2020.00235

DOI=10.3389/fgene.2020.00235

ISSN=1664-8021

ABSTRACT=Competing endogenous RNAs (ceRNAs) can regulate each other by competing binding microRNAs they shared. It is a vital post-transcriptional regulation mechanism and plays critical roles in physiological and pathological processes. Current computational methods for identifying ceRNA pairs are mainly based on expression correlation of ceRNA candidates and the number of their shared microRNAs, without considering the sensitivity of the correlation to the expression level of their shared microRNAs. To overcome this limitation, we introduce liquid association (LA), a dynamic correlation measure, which can evaluate the correlation sensitivity of ceRNAs to microRNAs, as an additional factor for ceRNA detection. To this end, we firstly analyze the effect of LA on detecting ceRNA pairs. Then, we propose a LA-based framework, LAceModule, to identify ceRNA modules by integrating the conventional Pearson correlation coefficient and dynamic correlation LA with multi-view non-negative matrix factorization. Using breast and liver cancer datasets, experimental results demonstrate that LA is a useful measure in detecting ceRNA pairs and modules. We find that the identified ceRNA modules play roles in cell adhesion, cell migration and cell-cell communication. Our results also show that ceRNAs may represent potential drug targets and markers for cancer treatment and prognosis.