...getting out what I want...
Tue Apr 11 09:37:54 EST 1995
przemko at reks.uia.ac.be (Przemko) wrote:
>I am familiar with both versions of SRS: command line and HTTP. During my work
>with it it became obvious that the features are not identical and that, in fact,
>only combination of the two would allow me to get what I want from that program.
>The problem that I face is a simple one:
>- get out all sequences that contain a signal peptide- only the ACTUAL feature
>- put it in GCG and do some things on it.
>The troubles are (for command line interface)
>- if I do something wrong (like asking FEATURE and PIR- it has no feature field)
yes that is true ...srs3.1 supports the PIR feature table and srs4.0 doesn't
this is because the feature information in PIR is rather scarce and sometimes
difficult to read (for a program) ...but in the near future the PIR feature
will be fully supported again
>the program mercilessly crashes (core dump etc.)
>- I cannot do q1-organism and q2-feature
peter rice has already explained how to do this on the command line
with the WWW server you just specify the words in the organism and feature
fields and then select "sequenceFeature" instead of the default "entries"
in the second last line of the query form ...click the ?! there to get
more information on that
>- I cannot do AllText search
this is a new feature in srs4
>- some other minor things
>For the HTTP interface:
>- I can do all above (except for the q1-organism and q2-feature)
>- BUT I cannot save my results in a file. What I want is all my seqs- FEATURES
>ONLY- as a file. One can do it in the command line interface but not HTTP.
you can do it with the srswww too ...do the query as described above and
then, on the next page click "view complete entries" ...this gives
you the list of found features plus some but not all of the parent entry's
Thure Etzold | EMBL
E-mail: etzold at embl-heidelberg.de | Postfach 10.2209
Tel: (49) 6221 387529 | 69012 Heidelberg
Fax: (49) 6221 387517 | Germany
More information about the Bio-srs