[Biopython-dev] Re: [BioPython] missing databases in Config.py of EUtils

[Michiel Jan Laurens de Hoon]

Hi Yu Huang,

I am now going through recent bugs and patches for Biopython for the upcoming 
new release of Biopython. I am not very familiar with EUtils. Can you give a 
small code example that uses your patch?

--Michiel.

Yu Huang wrote:
> Hi
> 
> File location: Bio/EUtils/Config.py. It misses some databases offered by 
> NCBI.
> I used two databases, unigene and gene. So I added them to the file.
>  > '''
>  > Added by yh
>  > '''
>  > databases.UNIGENE = _add_db(DatabaseInfo("unigene", 1))
>  > databases.GENE = _add_db(DatabaseInfo("gene", 1))
> 
> It's great that biopython now has the debian package!
> Yu
> 
> 
> ------------------------------------------------------------------------
> 
> """Configuration information about NCBI's databases"""
> 
> # Used to figure out if efetch supports the start, stop, strand, and
> # complexity fields
> PUBLICATION_TYPE = 0
> SEQUENCE_TYPE = 1
> 
> # Map from database name to database type
> class DatabaseInfo:
>     """stores NCBI's name for the database and its type"""
>     def __init__(self, db, dbtype):
>         self.db = db
>         self.dbtype = dbtype
> 
> class DatabaseDict(dict):
>     """map from name to DatabaseInfo for that database name
> 
>     Entries are also available through attributes like PUBMED,
>     OMIM, and NUCLEOTIDE.
>     """
>     def gettype(self, db, dbtype = None):
>         """Given a database name and optional type, return the database type"""
>         if dbtype not in (None, SEQUENCE_TYPE, PUBLICATION_TYPE):
>             raise TypeError("Unknown database type: %r" % (dbtype,))
>         if dbtype is None:
>             dbtype = self[db].dbtype
>         return dbtype
> 
> databases = DatabaseDict()
> 
> def _add_db(x):
>     databases[x.db] = x
>     return x.db
> 
> # XXX Try these
> # <option value="structure">Structure</option>
> # <option value="pmc">PMC</option>
> # <option value="taxonomy">Taxonomy</option>
> # <option value="books">Books</option>
> # <option value="geo">ProbeSet</option>
> # <option value="domains">3D Domains</option>
> # <option value="UniSts">UniSTS</option>
> # <option value="cdd">Domains</option>
> # <option value="snp">SNP</option>
> # <option value="unigene">UniGene</option>
> # <option value="popset">PopSet</option>
> 
> databases.PUBMED = _add_db(DatabaseInfo("pubmed", 0))
> databases.OMIM = _add_db(DatabaseInfo("omim", 0))
> databases.JOURNALS = _add_db(DatabaseInfo("journals", 0))
>                 
> databases.GENOME = _add_db(DatabaseInfo("genome", 1))
> databases.NUCLEOTIDE = _add_db(DatabaseInfo("nucleotide", 1))
> databases.PROTEIN = _add_db(DatabaseInfo("protein", 1))
> databases.POPSET = _add_db(DatabaseInfo("popset", 1))
> databases.SEQUENCES = _add_db(DatabaseInfo("sequences", 1))
> 
> '''
> Added by yh
> '''
> databases.UNIGENE = _add_db(DatabaseInfo("unigene", 1))
> databases.GENE = _add_db(DatabaseInfo("gene", 1))
> 
> # Someday I want to make it easier to get a given format.  I would
> # rather not have to specify the retmode/rettype pair, but I don't
> # know what people want from this feature, so skip for now.  Plus,
> # it's harder than I thought.
> 
> ##class FormatInfo:
> ##    def __init__(self, name, retmode):
> ##        self.name = name
> ##        self.retmode = retmode
> ##        self.rettype = rettype
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> BioPython mailing list  -  BioPython at biopython.org
> http://biopython.org/mailman/listinfo/biopython

-- 
Michiel de Hoon, Assistant Professor
University of Tokyo, Institute of Medical Science
Human Genome Center
4-6-1 Shirokane-dai, Minato-ku
Tokyo 108-8639
Japan
http://bonsai.ims.u-tokyo.ac.jp/~mdehoon