This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How To Parse The Xml From Entrez Db=Protein Using Biopython

I want to obtain a proteins gi no# along with its cd root no# and taxid from a proteins NP_xxx. 1 (id#) when I use the Biopython's entrez as following

 handle= Entrez.efetch(db='protein', id='NP_000368.1', retmode='xml')
 record= Entrez.read(handle)

the record is a list-elements object.and looks very complicated and I somehow can not really locate the "Tag" or "Key"??.How i parse the output to obtain the information I interested in?

BTW where I can see the overview for which DTD information has been retrieved from the ncbi?

biopython entrez protein xml
• 5,415 views • link
written by heath • heath

what do you mean with the "tag" or the "key" ?

• 0 views • link
written by Pierre Lindenbaum Pierre Lindenbaum

if the rettype change to "gb" 

handle= Entrez.efetch(db='protein', id='NP_000368.1', rettype='gb')
In [26]: ret=SeqIO.read(handle,'genbank')
In [27]: feats=set()
In [28]: for feat in ret.features:
   ....:     feats.add(feat.type)
   ....:    
In [29]: feats
Out[29]: set(['source', 'Protein', 'CDS', 'Region', 'Site'])

Keys. Tag might not be the accurate words but i meant the identities label/Tag name, I need to extract the Gi and CD information

• 0 views • link
updated by Istvan Albert Istvan Albert • written by heath • heath

why do you need to parse the XML manually? When above you already have BioPython reading the file

• 260 views • link
written by Istvan Albert Istvan Albert

eh...i am not sure if.Biopython is able to pass the xml from db='protein'. biopython should be able to parse most of the xml from entrez but not each one as long as I know... i can somehow put the above information into SeqIO.read(), and convert it into the Bio.SeqRecord but i can only have following informations.

 In [23]: from Bio.Seq import Seq

 In [24]: from Bio.SeqRecord import SeqRecord

 In [25]: from Bio.Alphabet import IUPAC

 In [26]: record = SeqRecord (rec2)

In [27]: print record
ID: <unknown id>
 Name: <unknown name>
 Description: <unknown description>

Number of features: 0 SeqRecord(seq=Seq('MSGGPMGGRPGGRGAPAVQQNIPSTLLQDHENQRLFEMLGRKCLTLATAVVQLY...WDD', IUPACProtein( )), id='NP000368.1', name='NP000368', description='wiskott-Aldrich syndrome protein [Homo sap iens].', dbxrefs=[])

but I need the protein gi along the protein_cdd information

• 0 views • link
written by heath • heath

I am confused. If you want a SeqRecord from a Biopython parser, just ask Entrez for a GenBank format file (as you have shown). Currently the SeqIO parsing framework doesn't handle the equivalent XML file - it could be done but appears to add relative little benefit in terms of new functionality.

• 29 views • link
written by Peter Peter

Thanks a lot for replying and sorry for confusing.. It does not really matter the retmode/rettype is 'xml' or'gb' ..I think at the moment I should simplify the question into "how i use Refseq id of a protein to obtain its Gi and CDD information."

• 0 views • link
written by heath • heath

I think that makes more sense and please post it as a new question

• 0 views • link
written by Istvan Albert Istvan Albert

0 answers

No answers yet.

Log in to answer this question.

More posts like this