1
$\begingroup$

I have a dataset composed on numerous RefSeq transcripts with names of the form NM_###### - essentially standard annotated mRNA accession numbers. I notice that numerous transcripts which have different sequences (different start and stop sites in a gene) all have the same NM_#### accession number. The interpretation is that they are all from the same gene, but this makes no sense to me. Why should different transcripts with different sequences have the same name? I understand this identifies their gene of origin, but they are still different transcripts.

Here is an example from the .bed file I'm using. Note they have different start and stop locations and even have different strand notations, but all have the same name.

chr     start        stop       name            score   strand
chrX    3441730     3443690     NM_001085545    0       -   
chrX    3700232     3702192     NM_001085545    0       +   
chrX    3750925     3752885     NM_001085545    0       -   
chrX    31336368    31338328    NM_001085545    0       +
Improve this question
$\endgroup$
7
  • $\begingroup$ Can you provide a few examples? If multiple entries have same NM_### accession but different sequences, could that be because they are different versions? You can view the entire version history of any NM_### accession on the NCBI pages. $\endgroup$
    – vkkodali
    Jun 10, 2019 at 5:37
  • 1
    $\begingroup$ What you're describing is almost certainly a mistake in your data. Different sequences don't have the same refseq IDs. Please give us an example. $\endgroup$
    – terdon ♦
    Jun 10, 2019 at 8:01
  • $\begingroup$ I added an example of what I see in the .bed file. Note that I misspoke a bit - I don't have the actual nucleotide sequences, but these all have different genomic start and stop sites, so they are effectively different sequences. $\endgroup$
    – KirkD-CO
    Jun 10, 2019 at 13:34
  • $\begingroup$ Are they the exons? $\endgroup$
    – Emily_Ensembl
    Jun 10, 2019 at 15:38
  • $\begingroup$ The current ensembl entry for that transcript is the only entry for that gene, it has a single exon spanning 3.70 to 3.72 Mb. So what is that entry at 31 Mb supposed to represent? $\endgroup$
    – swbarnes2
    Jun 10, 2019 at 16:08

1 Answer 1

Reset to default
4
$\begingroup$

No, this is the same sequence. A single RefSeq accession will correspond to one sequence. However, nobody said that sequence needs to be unique in the genome. The specific sequence you are talking about, NM_001085545.1, is a mouse transcript which can be aligned to many different regions of the mouse's chromosome X.

This is your sequence:

>NM_001085545.1
agagaagtctaggcaggactggcagactacagcaatgcctcaaccaagaagcctctgcga
gagcttgttctaggaattttcatcaatggggcttttagtcagcagggtcttgagatgcag
ggattccagtctgttagagccacagccagaagccatagccggagccagctacattcctgg
cagtcgcaagcgaaaaagaaacagtttggaggagttggcaacaagttctaatgttcatgg
acctcaaaaccagggaatgtatccacatcaagttctcaactacatctactggaaaagggt
taagatctcatctaatgatgcttatcaaaacttatttttggacgggcatgatagcgacat
taaaatccgtgctctgggaaaaacatggtgtttacacaaagtatttttatgtcagtcagg
ctactttgctaacattctcaaaggtacttggagagaatcacaccatggtgttataaatct
gatcattaagaatgaggatattgatacccgatctctgcattttgtgtttggtgctttgta
cacggatgcggatttgtcaataacacctctggaagttcctcaagttttggcagcagcatg
cctgcttcgggtggatcaagtaattcagcagtgtgaaggaatcatgaaagaaactatcaa
caggaacactgtgtgctcctattatttggcagcagaaacctatagattaaaagctgtaaa
gacgagatgctttgaatggcttctttgcaatttgatggtacatccaagtgtggcacttta
caaggaagtagacttgaagttgatgtatcttctagcactgtcttctgacttactagtcat
gcaaaaggagattgatgtatataccacactaaaaatatggatgttcctttatcttaatcc
atgctggaacggaaccatgaaacagcttttacaacacgcaaacaactggctttccaccca
catggcatatgttgataacatcagttttcttgaaagtgaagaaggactaatatttcaacc
agtgtttaaaaagctgagatttcagcacattatctgtgacttgacttccacaactattct
tgaacaagatcgactaatacctatggcatggttgtcacccatttacaaacaacagtggtt
gactttgctgcgaacacaagaatatggggtaattggaccacaagttatcaatgaacaaga
acttgaagaatgcaccatgaggtgtggtacaatgatccccaaggatggaagatatacttg
gaagtggtcagttggacgacttggctttcctttacgtgtgacctttaccaggcagtgtgt
aattttaaggcaacggtgtcagaggtgtgatggttctgcttgccacaaccatatccgaaa
tgtcatattcagaataactttggtgtgttttgattccaacaaaagagtaactttcagaaa
gacaacaggttataaaatcctcacctttgaatataaggaggagcaaattgtaatgaaatt
ggatagtgatgttctaaccttccctatgtgtatattctgcaatttcctttttgtaaacct
aggaaatgcagaaaacaagtaatcttatcaattattagactgtaagtattttacaagtat
tgaatggttcatttaatatgaaggctgtgtaatgactacagtgaagaagaccaaaaacaa
attcaatattctttggagtctagtaccaccagaagtgactgcttccctccatcaccttat
ttccatgtatttctgaagaaaacctttacttccaacaaagtatttgtcaaatcaaataaa
cctaatcttctgatttaaatttaccatcgagatccaagcttggaatggaaataagagact
atatttgcaaggtaaaacatctcaatcaatttgttaagtagatttggccctagctaaaat
tgagacaaaataaaatcatgccattgttgaaactgttgtg

If you BLAT it against the mouse genome using the UCSC BLAT tool, you will see many hits, most of them almost perfect matches and some of them overlapping the regions given in your bed file:

blat results

Your bed file matches the following hits in the image above:

NM_001085545.1  1954     1  1960  1960    99.9%  chrX   -     3441731   3443690   1960
NM_001085545.1  1960     1  1960  1960   100.0%  chrX   +     3700233   3702192   1960
NM_001085545.1  1954     1  1960  1960    99.9%  chrX   -     3750926   3752885   1960
NM_001085545.1  1954     1  1960  1960    99.9%  chrX   +    31336369  31338328   1960

I don't know why your bed file has all of these various matches for NM_001085545.1, since you haven't told us what the bed file is supposed to be doing. A bed file is just a collection of regions. For all I know, that bed file is just "a few hits for sequence NM_001085545.1 in the genome". There is no reason to assume that each region in a bed file corresponds to a gene.

So there's nothing wrong with your file, you're just dealing with a non-unique sequence from the mouse genome. And you don't have multiple sequences with the same accession, you have one sequence, one accession, but multiple hits for that sequence in the genome.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.