Software: miRNA analysis using eRNA

Abstract: The pipeline of miRNA analysis in eRNA named identification of known miRNA is used for identification of miRNA. 

This pipeline includes 4 steps: raw data reading, adapter removal, sequence alignment, and reads counting (Fig. 1). Click the run button to trigger running once all related parameters setup are finished.

Fig. 1 GUI of miRNA analysis.

Raw data reading

Parameters of raw data reading pipeline are setup in this window (Fig. 2).

• Sequencing direction: Default is single-end sequencing (R1), and paired-end sequencing is forced for mRNA analysis.
• Quality control: Quality score of a given base is defined by Q=-10*lg(e) where e is the estimated probability of the base call being wrong. Thus, a higher quality score indicates a smaller probability of error. A 100-base read matches 100 Q values in Illumina FASTQ format. A quality value Q is an integer mapping of p value of sequencing quality (approximately p < 0.05 equivalent Q > 13). Default cutoff of Q value is 13 in eRNA.
• QC compression: The number allows to compress Q values export (*.QC files) in o
  rder to save hard disk room and speed up QC graphing.

Fig. 2 GUI of raw data reading.

Adapter removal

Parameters of adapter removal pipeline are setup in this window (Fig. 3).

• Adapter sequences: 5’-end and 3’-end adapter sequences input.
• Adapter matching: eRNA identify adapter sequences based on 8 exact matching or 12 matching read sequences with at most one mismatched base allowed.
• Minimum query length: Sequences less than the minimum query length after adapter removal will be removed from sequence alignment.

Fig. 3 GUI of adapter removal.

Sequence alignment

The third-party aligner (the default is Bowtie1) is required in this module ().

• Options in Bowtie: eRNA only provide the options of Bowtie1 on -m, –n/-v, -l, and -a/-k/-a --best –strata. See the details about Bowtie1 from the web site (http://bowtie-bio.sourceforge.net/index.shtml).

Fig. 4 GUI of sequence alignment.

Users can choose the different methods of sequence alignment (Fig. 5).

• Separate alignment: Multiple reference sequences for alignment are allowed. Read sequences from each miRNA sample are separately aligned with those selected references one by one.
• Iterative alignment: Read sequences for each miRNA sample are aligned with those selected references in order. Those unaligned read sequences in the previous alignment are the input of the next alignment. Note that the order of iterative alignment results in different alignment result.

Fig. 5 Comparison between separate and iterative alignment.

Reads counting

Qualification is done in this step (Fig. 6).

• Reads background: Any read sequences determined by sequencing below the number of reads background is marked as none-detection and to be neglected.
• Normalization of transcriptional level: Transcriptional level of a miRNA can be measured by its read counts (RC), or the read counts normalized by the mappable reads counts (MRC) or by the total read counts (TRC) of a miRNA library.

Fig. 6 GUI of reads counting.

Writing data: 2013.12.05, 2015.02.06