refseq accession alias table

Question

[jacobsen.jeremy]

I've noticed that some uniprot accessions link to multiple NM_ refseq nucleotide accessions. For instance (Q8NDI1-3 -> NM_001142615, NM_001142616). This suggests to me that there are multiple refseq accessions which can link to the same sequence (redundancy). If this is the case, is there a way to retrieve a table of aliases? For instance, line 1 would be: NM_001142615, NM_001142616.....

Thanks!

jeremy

Answer 1

[Pierre Lindenbaum]

Under http://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/idmapping/

$ curl -s "ftp://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/idmapping/idmapping.dat.gz" | gunzip -c | grep "Q8NDI1"
(...)
Q8NDI1-3    RefSeq_NT    NM_001142615.2
Q8NDI1-3    RefSeq_NT    NM_001142616.1
(...)

[jacobsen.jeremy]

Thanks Pierre. I'm actually hoping to get this data from ncbi, or another source that contains all aliases for NM_ or NR_ accessions. I'm looking to compare proteomics data to RNAseq, so my hope is that if I have a uniprot_acc->refseq_nuc mapping, I can lookup the matching entry in my RNAseq results. It is sometimes the case that I have a uniprot_acc->refseq_nuc mapping, but cannot find the refseq_nuc match in the cufflinks output due to versioning differences between the gtf annotation and the refseq accession that is in uniprot.

Say for instance that (NM_123==NM_234==NM_456) and that uniprot has Q9876->NM-123, but Cufflinks reports an FPKM for NM_234. If I had a list Q9876->(NM_123,NM_234,NM_456) then this could be resolved.

Thanks again.

-Jeremy