Hi Sachin, 

you are right: the gold-standard for validating differentially expressed miRs is by qRT-PCR.

It seems to me that 160 miRs are too many. My suggestion is that you start validating few of them (lets say 5) choosing both, with high, medium and low FC values and see what the PCR says…If all of them are validated you are still facing the same problem, but I would bet than most of them are false positives….

Apart from choosing those with highest FC values, another selection criteria you might use could be based on a gene ontology and pathway analysis study… For this, you have several open software available, including DAVID; WikiPathways or CytoScape´s ClueGO application…

Good luck!

Because a huge amount of miR over and down expression you can use the criteria of the amount of fold change, for exemplo, miR with more or less than 10, 15, 20  fold change( it depend of your objective. Choose them and validade in at least a representative statistical number that you can discuss with a speciallist in this field

Thanks Javier and Mario for your suggestions. I am already thinking on the same lines. Problem is that ordering TaqMan miRNA assay for 10 miRNAs for validation and if they fail, orderning for next 10 miRs, so this cycle of validation will be very costly. I was thinking of reducing the chances of failure atleast with 70-80% of miRNAs choosen for validation. 

Hi Sachin,

I think you can reduce the miRNA count for validation if you can functionally enrich them.

So for that you can group the Differentially  expressed miRNAs as highly expressed and  low.Then do some target enrichment (miranda) against Differentially  expressed gene set of your disease of interest. There you will find some set of miRNAs enriched with some functionally enriched targets relevant to the disease. I guess fromthere you can screen some miRNAs  to validate them.

Thank you Jyoti. Seems I need to find some bioinformatician for this.