using NGS, what is the best miRNA expression normalization method?

[Giorgio C]

Hi All,

i need to check miRNA differential expression between two 454 run lanes. For this, i summed up all the reads that aligned to each miRNA. and i now have 2 tables that specify each miRNA and the number of reads that aligned to it.

Obviously a normalization needs to be done when comparing these two tables. my question is: normalization to what? should i use the total number of reads (mapped and unmapped) produced from each lane? or should i use the number of mapped reads? or the number of uniquely mapped reads?

[aggp11]

Hi All,

i need to check miRNA differential expression between two 454 run lanes. For this, i summed up all the reads that aligned to each miRNA. and i now have 2 tables that specify each miRNA and the number of reads that aligned to it.

Obviously a normalization needs to be done when comparing these two tables. my question is: normalization to what? should i use the total number of reads (mapped and unmapped) produced from each lane? or should i use the number of mapped reads? or the number of uniquely mapped reads?

Giorgio,

Instead of normalizing the reads yourself, you could try using DESeq, to identify the differentially expressed miRNAs.

I am guessing, that since you want to check for differentially expression you have two different conditions? For DESeq you'll need a file like this:

miRNA Condition1/Lane1 Condition2/Lane2
mir-1 23 45

where, condition1 and Condition 2 columns have the number of reads that mapped to the corresponding miRNA.

I hope this helps.

[Giorgio C]

Hi,
first thanks a lot for your reply. Precisely i have two set of miRNAs 1st from leaf | 2nd from Flower both from the same plant at the same time, condition etc.etc...

I've read about DeSeq and i'm trying to use it but with not success for now ;

However you know if the DeSeq takes into account that 1st set has less initially reads than 2nd set? Or it isn't of any importance because it uses only mapped reads ??

[aggp11]

Interesting... I don't think that DESeq worries about the initial number of reads as you are only providing what is known and has been identified in both your samples.

I recently did a miRNA analysis myself using DESeq and one of my samples had more reads as compared to the other and I think DESeq normalizes based on the mapped reads.

Thanks,
Praful

[Giorgio C]

Ok i hope DeSeq will work.

Thank to you Praful

Cheers