How to process the raw reads from miRNA-seq

Question

[smallfish]

Hi,

Many studies reported the peak length of miRNA of 22 nt in fishes. However, I obtained the peak length of 24 nt. Is there any mistakes in the data processing?

The raw reads were obtained by illumina hiseq xten sequencing, and were processed with the program cutadapt to trim the adaptor sequence (TGGAATTCTCGGGTGCCAAGGAACTCCAGTCACATCAC) from the 3' end. Reads with poor quality, or shorter than 18 nt were also removed.

Should I cut one base for per read in its 5' and 3'?

[GenoMax]

Some aligners should be able to soft-clip the bases on end. You want to use an ungapped alignment. If you try bbmap.sh from BBMap suite use these parameters: ambig=all vslow perfectmode maxsites=1000 for miRNA.

[Carlo Yague]

You can also check your reads with fastqc or similar before/after trimming. The "per base sequence content" will tell you what are the "extra" nucleotide at the 5' or 3' end.