Tracking miRNA precursor metabolic products and processing sites through completely analyzing high-throughput sequencing data

Abstract

The small non-coding important regulatory molecules, microRNAs (miRNAs), have been widely and deeply studied especially combining high-throughput sequencing technologies. Here, we attempted to track detailed miRNA precursor metabolic products and gain further insight into pre-miRNA processing by completely analyzing high-throughput sequencing data. Highly expressed miRNA precursors could be entirely covered by various short RNAs and small RNA fragments with a hierarchical distribution. miRNAs and some miRNA* regions were detected quite abundant short RNAs as expected, while other regions of precursors were found shorter RNAs or small fragments with fewer sequence counts. Furthermore, we developed a method to analyze relative expression levels of special RNA classes according to divergence of 5' and 3' ends, respectively. Generally, there were several quite abundant RNA classes from a given miRNA locus, which suggested dominant cleavage sites of Drosha and Dicer during pre-miRNA processing. Compared with 3' end, dominant cleavage site in 5' end always focused on a specific position, which ensured conservation of the identity of miRNA (5'-seed sequence, nucleotides 2-8). Overall, a comprehensive analysis of sequencing data can be used to track pre-miRNA metabolic products and mechanism of pre-miRNA processing and metabolism.