Functional in silico analysis

Since certain circRNAs may function as sponges sequestering miRNAs and, therefore, may be involved in the regulation of gene expression, it was interesting to explore the potential interactions between differentially expressed circRNAs in stroke and miRNAs. The circRNA/microRNA interaction was predicted with Arraystar’s home-made miRNA target prediction software based on TargetScan & miRanda [19], and the differentially expressed circRNAs were annotated in detail with the circRNA/miRNA interaction information. Then, overrepresented miRNAs (those linked to at least four differentially expressed circRNAs) were analyzed by DIANA-mirPath v.3 software [20] to predict the underlying pathways.

CircInteractome tool provided a list of miRNAs potentially targeted by the circRNA of interest and mapped binding sites for RNA-binding proteins (RBPs) on it [21]. Next, DIANA-mirPath v.3 software [20] analyzed the pathways in which the outcome miRNAs were involved based on TarBase v7.0, microT-CDS v5.0 and TargetScan. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses were used to predict cell signaling pathways and functions related with the set of outcome miRNAs.

Moreover, in order to assess the interactions betweem the RBPs predicted to bind a particular differentially expressed circRNA, protein list was uploaded to the Search Tool for the Retrieval of Interacting Genes (STRING) tool (Version 11.0) [22] and filtered for interactions of high confidence (score > 0.7) and PANTHER (Protein ANalysis THrough Evolutionary Relationships) Classification System (Version 14.1) [23]. Only those terms with a FDR-corrected p-value < 0.05 were reported.