How to figure out adapter sequence to use for trimming old miRNAseq data

Question

[hsothers]

I'm trying to analyze some old miRNAseq data and I'm having a hard time getting the reads trimmed to the expected length. I ran FastQC on the fastq files and despite all the reads being the same length (55bp), they all had <5% adapter content.

I've tried trimming with CutAdapt using the adapter sequences from Illumina, and with different settings for CutAdapt. However, none of it worked, the majority of the reads were still ~55 bp. I also checked a few of the full length reads in mirBase to see if the data actually needed to be trimmed, and none of them yielded any hits. So, I'm assuming they do, but I could be wrong since this is my first time starting from fastq files for RNAseq data..

How do I figure out what adapter/sequence needs to be trimmed from my data or if it needs to be trimmed at all before alignment?

Other relevant info:

• These were sequenced ~10 years ago.
• The library was made with either the Qiagen or Illumina kits for small/miRNA.
• It's paired end.
• Expected read length is ~22 bp.

Answer 1

[GenoMax]

fastp should be able to figure out the adapter. Use R1 (R2 is not going to add any additional info).

You can probably use bbmerge.sh from BBMap suite as well: Identify adapter sequences for trimming from Illumina paired end fastq files

If you convert the R1 to fasta and view the sequence (kind of look like a multiple sequence pileup) the adapter should become easy to spot. Not every read will have the adapter and it may not exactly be starting at 23rd base.

Try something like

reformat.sh -Xmx4g in=read1.fq.gz out=stdout.fasta reads=150 | grep -v "^>" | less -S

[hsothers]

Thank you!