miRDeep*: an integrated application tool for miRNA identification from RNA sequencing data

Abstract

miRDeep and its varieties are widely used to quantify known and novel micro RNA (miRNA) from small RNA sequencing (RNAseq). This article describes miRDeep*, our integrated miRNA identification tool, which is modeled off miRDeep, but the precision of detecting novel miRNAs is improved by introducing new strategies to identify precursor miRNAs. miRDeep* has a user-friendly graphic interface and accepts raw data in FastQ and Sequence Alignment Map (SAM) or the binary equivalent (BAM) format. Known and novel miRNA expression levels, as measured by the number of reads, are displayed in an interface, which shows each RNAseq read relative to the pre-miRNA hairpin. The secondary pre-miRNA structure and read locations for each predicted miRNA are shown and kept in a separate figure file. Moreover, the target genes of known and novel miRNAs are predicted using the TargetScan algorithm, and the targets are ranked according to the confidence score. miRDeep* is an integrated standalone application where sequence alignment, pre-miRNA secondary structure calculation and graphical display are purely Java coded. This application tool can be executed using a normal personal computer with 1.5 GB of memory. Further, we show that miRDeep* outperformed existing miRNA prediction tools using our LNCaP and other small RNAseq datasets. miRDeep* is freely available online at http://www.australianprostatecentre.org/research/software/mirdeep-star.

Figure 1.

miRNA prediction strategy for miRDeep, miRDeep2 and miRDeep*. (a) miRNA prediction pipeline for miRDeep*. Step 1 involves mapping sequences to the genome to produce a SAM file. In step 2, aggregated RNAseq reads are assessed for pre-miRNA secondary structure potential. (b) The two different miRNA prediction strategies (case 1 and case 2) for miRDeep. (c) and (d) Modified miRNA prediction strategy for miRDeep2 and miRDeep*.

Figure 2.

The input/output interface for miRDeep*. Parameters are customizable for the genome build, use of the original mirDeep algorithm (miRDeep), filtering out platform-specific adapter sequences (Adapter), the miRNA length (miR Length), the minimum read quality (min phred), the maximum number of loci mapped for a particular read (max multimap), the minimum number of reads (min reads) and the minimum prediction score (min score). The location of the mature miRNA and the number of reads relative to the pre-miRNA secondary structure can also be displayed by simply clicking on the miRNA of choice. On the pop-up window, target genes can be predicted for current sequence by clicking ‘Target genes’. The sequence in the top textbox can be replaced and click ‘RNA structure’ to regenerate RNA secondary structure. Note that mature sequence needs to be in upper case.

Figure 3.

(a) Vertical axis represents the log scores of predicted miRNAs obtained from miRNA prediction algorithm. The horizontal axis lists predicted miRNAs in descending order of score. Green lines show the predicted miRNAs that are known from miRBase. The red lines show the predicted miRNAs not listed in miRBase (i.e. novel miRNAs). The vertical axis is log scale, because the score difference across predicted miRNAs is huge. Since miRanalyzer, miRTRAP and MIReNA do not provide scores, the vertical axes represent the number of reads in the predicted miRNA. The number of predicted miRNAs shown in the figure is reduced to match the other algorithms. (b) Vertical axis represents the log fold changes after anti-dicer induced. The horizontal axis lists predicted miRNAs in descending order of folder changes.

Figure 4.

Validation of miRDeep*-predicted novel miRNA. (a) Validation was carried out using stem-loop TaqMan PCR for four novel miRNA (novel 1–4), and hsa-miR-29a and hsa-let-7a-2, which served as positive controls. Small RNAseq was performed on RNA that was extracted from LNCaP prostate cancer cells that were treated with ethanol (mock) or androgen (1 nM R1881) for 48 h. Represented is the mean expression ± SE from three independently treated cells. (b) Representation of the pre-miRNA secondary structure predicted by miRDeep*, including the location of the mature miRNA (red nucleotides). Also shown are the locations and number of RNAseq reads (lines above and below the predicted pre-miRNA). (c) Qualitative RT–PCR showing expression of novel 1, 2, 3, and 4 pre-miRNA. No reverse transcriptase (−RT) controls were included to ensure that genomic DNA (gDNA) was not amplified. gDNA and a negative template (H₂O) were included as positive and negative controls, respectively.