RefSeq ID to TaxID?

Question

[bird77]

I have a number of RefSeq IDs (like this here NZ_LFWC01000004.1) and I would like to get the tax IDs for the species.

Is there a way to automatize that in Bash, Python or R?

[erwan.scaon]

If you have a very long list of RefSeq IDs, you might want to do a local search vs "accession2taxid" files, see : Biostars

[bird77]

Wonderful, that is exactly what I was looking for. Thank you so much.

[Bastien Hervé]

Python and entrez module ?

Answer 1

[Sej Modha]

Using NCBI Unix eutilities:

esearch -db nucleotide -query "NZ_LFWC01000004.1"|esummary|xtract -pattern TaxId -element TaxId

[bird77]

Any idea about this error?

$  esearch -db nucleotide -query "NZ_LFWC01000004.1"|esummary|xtract -element TaxId

ERROR: No -pattern in command-line arguments

xtract seems to cause the error.

[Sej Modha]

Try this:

esearch -db nucleotide -query "NZ_LFWC01000004.1"|esummary|xtract -pattern TaxId -element TaxId

[bird77]

ah, wonderful, thank you very much. :-)

Answer 2

[Bastien Hervé]

In python :

from Bio import Entrez
from Bio import SeqIO
key_list=['NZ_LFWC01000004.1'] ###Add all your IDs

for key in key_list:
    Entrez.email = "myemailaddress"
    handle = Entrez.efetch(db='nucleotide', id=key, rettype='gb')
    record = SeqIO.read(handle,'genbank')
    if record.features[0].qualifiers['db_xref'][0].split(":")[0] == 'taxon':
        print(record.features[0].qualifiers['db_xref'])[0].split(":")[1]

[Sej Modha]

I guess one has to be careful with the db_xref tag, it can often contain identifiers linking to other databases such as UniProtKB.

[Bastien Hervé]

Yes right, I added a condition to keep 'taxon' id from NCBI's taxonomic identifier only