the vanishing index mystery

Thure Etzold etzold
Mon Jun 19 12:56:36 EST 1995


gbottu at ben.vub.ac.be (Guy Bottu) wrote:
>After upgrading from srs 4.4 to srs 4.5, I have observed something strange :
>the "Sequence" field of TFSITE is not indexed anymore. Yet the srsupdate
>script contains a line :
>srsbuild TFSITE -c -f ' SEQ' -xdir 'SRSINX:' -odir 'SRSINX:' -env 'unix'
>executing this line gives no error message, even when parameter -w
>is appended, only, the indexes tfsite_seq.* are not made...

hi Guy,

yes you spotted a problem ...which is that two definitions for the 
data-field "Sequence" exist. One for SwissProt where the Sequence data-field
is the line starting with "SQ" and is for display only and the "Sequence"
data-field for Tfsite where the sequence should be really indexed.

The solution is to generate a new field type "DNASequence" definition.
So just add the following into the file "srsgeneral.sdl":

#srsfield /id=%DF_DNASEQ
          /shortname="SEQ" /name="DNASequence" 
          /build=@F_BLDKEY /itype=key

then change the pointer in transfac.sdl from @DF_SEQ to @DF_DNASEQ:

    #field /itype=key /ftype=@DF_DNASEQ /idtype=@ENTRY_ID
           /find=sequence /begstr="SE  " /nextstr="SE  "

after running
srssection
try 
srsbuild -d tfsite -f dnasequence

to test if the sequences are really extracted

regards
thure




===============================================================================
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 mailing list