converting mirDB target ids to ensembl ids

[bioBob]

Hi,

I have about 3000 miRNA targets from miRDB. I am needing to convert the target identifiers into Ensembl ids (mouse btw). About 2k of these convert using biomaRt in R. The other 1k I am not sure how to handle programmatically as I can't figure out what id type they are.

Most convert using either refseq_mrna (Refseq mRNA ID(s) [e.g. NM_001195244]) or refseq_mrna_predicted (Refseq Predicted mRNA ID(s) [e.g. XM_889253])

An example of one that doesn't convert: "NM_001042565". Looking this up in Ensembl via the main search I get "novel transcript (Mouse Transcript)". So I know it is there. I have taken the identifier and tried to convert it using Ensembl's BioMart using the filter option with most of the gene/transcript looking options and do not get anything back.

What filter option should I use to get all the miRDB target id's to convert to Ensembl IDs???

Thanks,
Bob

[yueluo]

Took a quick look at both Ensembl and Ncbi considering the transcript "NM_001042565". I tried its gene "Wsb1" to get all the transcript information.

NM_001042565.3 = ENSMUST00000017821 (Wsb1-001)
NM_019653.3 however, has no corresponding Ensembl transcript.

This would probably be why NM_001042565 doesn't get converted, since biomart would not contain an id-id mapping information for such transcript.

Anyway, what does the original "miRDB target id" look like?

[bioBob]

Hi,

thanks for the reply. That is the target id. From miRDB:

mmu-miR-669f-3p NM_001042565 99.7800236967459

I suppose I could get around the ensembl id's by using gene name, but I still need a way to bulk convert the id's to gene names.