Abstract
Background In-spite of ubiquitous expression of DROSA/DICER, miRNA formation and maturation are highly spatiotemporal implying involvement of other factors in their biogenesis. Several key studies have elucidated functions of few other RNA-binding proteins (RBPs) in miRNAs biogenesis, making it necessary to look miRNA biogenesis models with fresh approach.
Results A comprehensive study of >25TB of high-throughput data revealed that various combinations of RBPs and their networks determine the miRNA pool, regardless of DROSHA/DICER. The discovered RBP and miRNA associations displayed strong functional alliances. An RBP, AAR2, was found highly associated with miRNAs biogenesis, which was experimentally validated. The RBPs combinations and networks were tested successfully across a large number of experimentally validated data and cell lines for the observed associations. The RBP networks were finally modeled into a XGBoosting-regression based tool to identify miRNA profiles without any need of doing miRNA-seq, which scored a reliable average accuracy of 91% on test sets. It was further tested across >400 independent experimental samples and scored consistently high accuracy. This tool was applied to reveal the miRNAome of Covid19 patients about which almost negligible information exists. A significant number of Covid19 specific miRNA targets were involved in IFN-gamma, Insulin/IGF/P3K/AKT, and Ub-proteasome systems, found in cross-talk with each other and down-regulated heavily, holding promise as strong candidates for therapeutic solution. A large number of them belonged to zinc-finger family.
Conclusion There are several RBPs and their networks responsible for miRNA biogenesis, regardless of DROSHA/DICER. Modeling them successfully can reveal miRNAomes with deep reaching impact.
Competing Interest Statement
The authors have declared no competing interest.
