How to measure miRNA expression from Illumina libraries

[foolishbrat]

I have a 51 single-end reads generated with MiSeq using NEBNext Multiplex Oligos for Illumina.

The data is obtained from from blood serum miRNA.

The sample sheet looks like this:

IEMFileVersion,4
Investigator Name,FB
Experiment Name,WT10104
Date,11/27/2013
Workflow,GenerateFASTQ
Application,FASTQ Only
Assay,TruSeq Small RNA
Description,
Chemistry,Default

[Reads]
51

[Settings]
ReverseComplement,0

[Data]
Sample_ID,Sample_Name,Sample_Plate,Sample_Well,I7_Index_ID,index,Sample_Project,Description
HS130333-1,,,,RPI3,TTAGGC,,
HS130333-2,,,,RPI4,TGACCA,,
HS130333-3,,,,RPI5,ACAGTG,,

So it contain 3 files:

HS130333-1.fastq
HS130333-2.fastq
HS130333-3.fastq

My question is, if I want to measure the miRNA expression,
should I combine all the 3 files above, then map them before computing the expression?

OR map each of the above 3 files individually and then average the expression?

[TiborNagy]

I suggest you to map individually, because you get more statistical power for further analysis.

[RedLightPanic]

Hi foolishbrat,

I agree with TiborNagy that the best option is to map them individually and the compute the miRNA expression separately.

On the other hand, I also suggest you to apply some type of normalization before going further on the analysis of miRNA expression. Here) you'll find a survey of normalization methods for small RNA-Seq data.

Hope this helps

[foolishbrat]

Thanks a million both.