Global analysis of gene-level microRNA expression in Arabidopsis using deep sequencing data

Abstract

MicroRNAs (miRNAs) regulate gene expression at the post-transcriptional level in eukaryotes. Exclusive focus on mature miRNA in most expression profiling efforts has prevented effective measurement of the expression of individual miRNA (MIR) genes. Using three sequenced small RNA libraries, we adapted miRDeep, which employs a probabilistic model of miRNA biogenesis, to analyze the miRNA transcriptome in Arabidopsis. We determined that less than 40% annotated MIR genes are expressed in shoot, root or inflorescence. We found that within paralogous families the expression pattern of individual genes correlates with the phylogenetic distance. Combining novel candidates identified in this study, we deduced the maximal number of expressed MIR genes. We further estimated the sequencing depth necessary to reach a near-saturated detection rate by curve fitting simulation. These results demonstrate that signature distribution of small RNA reads along the miRNA precursor is an effective model to profile MIR gene expression in Arabidopsis.