ABSTRACT
Identification of long and short RNAs, their processing and expression patterns have been greatly facilitated by high-throughput sequencing. Frequently, these RNAs act as guides for ribonucleoprotein complexes that regulate the expression or processing of target RNAs. However, to determine the targets of the many newly discovered regulatory RNAs in high-throughput remains a challenge. To globally assign guide small nucleolar RNAs to site of 2’-O-ribose methylation in human cells, we here developed novel computational methods for the analysis of data that was generated with protocols designed to capture direct small RNA-target interactions and to identify the sites of 2’-O-ribose methylation genome-wide. We thereby determined that many “orphan” snoRNAs appear to guide 2’-O-ribose methylation at sites that are targeted by other snoRNAs and that snoRNAs can be reliably captured in interaction with many mRNAs, in which a subsequent 2’-O-methylation cannot be detected. Our study provides a reliable approach to the comprehensive characterization of snoRNA-target interactions in species beyond those in which these interactions have been traditionally studied and contribute to the rapidly developing field of “epitranscriptomics”.
