miRNA target genes prediction.

Question

[anu014]

Hello Biostars!

Today I'm trying a new pipeline for miRNA prediction. I used 'miRDeep' for miRNA prediction & after that I used 'DESeq' for differential analysis. Now I'm wondering how to get target genes for these miRNA. There're several packages such as targetscan.Hs.eg.db, multiMiR, miRNATarget etc. Which one do you all suggest?

And what all downstream analysis can be done related to miRNA ?

Please do suggest. Thanks you all :)

Answer 1

[Kevin Blighe]

I have recently used miRNAtap, which produces an aggregate score from up to 6 mir target databases and ranks the predicted targeted genes.

This can then be used in conjunction with topGO and KEGGprofiler in order to perform gene enrichment and pathway analysis of the predicted target genes.

Further information, take a look here: A: Identifying miRNA, RNA-binding sites in UTRs

Kevin

[anu014]

It's pretty cool. One of the best package I found today. Thank you.

But I am still not fully satisfied, its manual says : 'Currently predictions are available for Homo sapiens, Mus musculus and Rattus norvegicus (the last one through homology translation)'.

Can it be customized for different species too, somehow?

[Kevin Blighe]

On which species are you working? It can be that the data you require simply does not yet exist.

[anu014]

Right now I'm doing testing on Human data. But I've to work on Danio rerio in few days..

[Kevin Blighe]

Okay, you may get some ideas from this study:

• Screen of MicroRNA Targets in Zebrafish Using Heterogeneous Data Sources: A Case Study for Dre-miR-10 and Dre-miR-196

In additon, this database appears to have zebrafish:

• miRNA_Targets: A database for miRNA target predictions in coding and non-coding regions of mRNAs

Answer 2

[lieven.sterck]

Its' not really an R package so not sure if this one can be of use : TAPIR

but it has a standalone version as well (see the manual pages)

[anu014]

It's nice one, but it's limited to plants only.. I wonder can we customize it for different organisms too.

[lieven.sterck]

Good question. should somehow be possible I guess but you probably will have to go into the code and change a few things, and of course you will need the appropriate info (eg. typical miRNA length etc) to customize it