miRNA-seq alignment bowtie2

Question

[imaparna27]

I am trying to align miRNA-seq reads against miRBase reference using Bowtie2, but I am getting a very low overall alignment rate.

using Bowtie2:

11371657 reads; of these: 11371657 (100.00%) were unpaired; of these: 11371652 (100.00%) aligned 0 times 0 (0.00%) aligned exactly 1 time 5 (0.00%) aligned >1 times 0.00% overall alignment rate

using HiSat2:

11371657 reads; of these: 11371657 (100.00%) were unpaired; of these: 11371455 (100.00%) aligned 0 times 50 (0.00%) aligned exactly 1 time 152 (0.00%) aligned >1 times 0.00% overall alignment rate

What could be the possible reason causing difference in alignments? Why the overall alignment rate is this low? The read lengths are around 20nt (no adapters after quality check).

[Trivas]

It would be helpful if you could provide some more information such as your actual commands for both mappers and how you indexed the miRBase reference prior to mapping.

[imaparna27]

The index was built for mature miRNA obtained from miRBase22 using following command:

bowtie2-build mature.fa bowtie2_index

The following commands were used for both mappers: Bowtie2

bowtie2 --very-sensitive-local -x my_path/bowtie2_index -U my_path/my_file.fq.gz  -S my_file.sam

Hisat2

hisat2 -p 12 --dta -x /my_path/hisat2_index -U my_path/my_file.fq.gz
-S my_file.sam --un-gz my_file_unaln.fq.gz --summary-file my_file_sum.txt --met-file my_file_met.tsv

[imaparna27]

I have used both trimmomatic and cutadapt to remove the adapter sequence from the miRNA-Seq samples, and I am certain samples are of good quality, then why this less alignment rate?