Paired-end miRNA pairs to be used as replicates?

[FoRsUs]

Dear community,

I wanted to ask your advice on the following question:

I have 2 conditions(datasets) of illumina paired-end miRNA-seq without replicates. I was wondering if could be possible to use each pair as a replicate in order to quantify miRNA expression?

This question came to my mind because I was looking to analyze this data and could not find any pipeline well suited for paired-end miRNAs, do you know any available pipeline or have any advice also on this matter?

Thanks for your help!

[dpryan]

I assume you mean to ask if it's appropriate to treat the reads of a pair separately in both alignment and quantitation, treating read #2 as a biological replicate of read #1. If so, that's a really bad idea and would seriously inflate your false positives (if you did that, you would be vastly underestimating intra-group variance). Without actual replicates, it's best to just be honest about it and make a list with fold-changes for follow-up.

[FoRsUs]

Thanks for your answer, I didn't do it already I was trying to think what can I do when no replicates are available. Also because, I looked at some pipelines and and didn't find any which could manage paired-end data

[dpryan]

Yeah, since miRNAs are short and reads lengths continue to increase (they've been longer than mature miRNAs for a while), I expect there's little if anything gained from using paired-end data when dealing with miRNAs.