[Biopython-dev] Bio.WWW.ExPASy

[Michiel De Hoon]

Hi everybody,

The Bio.WWW reorganization has been completed for Bio.WWW.InterPro and
Bio.WWW.SCOP. Now I am looking at Bio.WWW.ExPASy.
Bio.WWW.ExPASy is used to access SwissProt, Prosite, and Prodoc records
through ExPASy. Because of the overlapping functionality in Bio.WWW.ExPASy I
now think it is better to keep this module together as Bio.ExPASy instead of
dividing it into Bio.SwissProt and Bio.Prosite.

Bio.WWW.ExPASy contains six functions:

get_prodoc_entry  Interface to the get-prodoc-entry CGI script.
get_prosite_entry Interface to the get-prosite-entry CGI script.
get_prosite_raw   Interface to the get-prosite-raw CGI script.
get_sprot_raw     Interface to the get-sprot-raw CGI script.
sprot_search_ful  Interface to the sprot-search-ful CGI script.
sprot_search_de   Interface to the sprot-search-de CGI script.

plus an internally used function _open. (I added the get_prosite_raw function
as it was missing from the existing Bio.WWW.ExPASy).
These functions call the _open function with the appropriate URL.

Inside the _open function, we have the following code:


    # Wrap the handle inside an UndoHandle.
    uhandle = File.UndoHandle(handle)

    # If the key doesn't exist, ExPASy returns nothing.
    if not uhandle.peekline():
        raise IOError, "no results"

Unfortunately, it is no longer true that ExPASy returns nothing for a
non-existing key. Nowadays, it returns an HTML web page with an error
message. The only exception is the get_prosite_raw function I just added;
here, ExPASy indeed returns nothing for a non-existing key.

So, the uhandle.peekline() will not fail even if the key is non-existing, and
the IOError will not be raised.

In this situation, I think it is better if the functions in Bio.WWW.ExPASy do
not attempt to raise an IOError, and to let the subsequent parsers handle
non-existing queries. The function get_prosite_raw could be an exception,
though I wouldn't be surprised if ExPASy one day decides to make its behavior
consistent with the other functions.

Any comments?

--Michiel.



Michiel de Hoon
Center for Computational Biology and Bioinformatics
Columbia University
1150 St Nicholas Avenue
New York, NY 10032