[BioPython] Does anyone use EZRetrieve?

[Peter]

On Thu, Nov 20, 2008 at 8:53 PM, Peter <biopython at maubp.freeserve.co.uk> wrote:
> On Thu, Nov 20, 2008 at 8:46 PM, Bruce Southey <bsouthey at gmail.com> wrote:
>> Hi,
>> Does anyone use EZRetrieve
>> (http://siriusb.umdnj.edu:18080/EZRetrieve/single_r.jsp) ?
>> This allows a user to retrieve a human, mouse or rat genome nucleic sequence
>> based on an valid identifier.
>>
>> I think that most of the functionality of Bio.EZRetrieve is already present
>> in Biopython and the genome sources appear to be 5 years old. For example,
>> it uses LocusLink that was discontinued March 2005.
>>
>> If so could you please let me know?
>
> Actually - could you let the whole mailing list know?  ;)
>
> Given nature of the database and the limited functionality this python
> code offers, if no-one is using Bio.EZRetrieve then it could be
> considered for deprecation.

I've seen no replies so I've marked Bio.EZRetrieve as obsolete in CVS
(and therefore for Biopython 1.50), and unless anyone speaks up it
will be deprecated in the release after that.

I'm not sure that the EZRetrieve data is that out of date (they may
have updated things since Bruce looked), but all the Bio.EZRetrieve
code does is fetch an HTML page and extract the FASTA formatted
sequence (ignoring any metadata or cross references).  In any case,
this kind of HTML "screen scraping" is fragile (liable to break when
the site gets a visual redesign) and is not explicitly condoned by the
service itself.

Peter