Keith Elliston keith_elliston at merck.com
Thu May 26 13:30:54 EST 1994

> I think that Keith's suggestion (for finding valine-less 50-mers or less) was
> practically brilliant in its simplicity, however, as Conrad and Mike stated, I
> don't think it works on version 7.3 (I know that it doesn't on the Solaris 
> version).  On our system, when you use the pattern "<~V{1,49}~V>" what appears
> to come up as matches are proteins that do not have a valine in their last
> 50 residues.  In fact, the program appears to convert the query to ~V{49,49}~V>
> (49 non-valines followed by a non-valine at the end of the peptide): 
 (etc. deleted....ke)

I re-ran my search, and there are no sequences longer than 50 residues in my 
output.  I took a look at my (VMS) version of findpatterns, and this is what
I have....


FINDPATTERNS.FOR;2    6-FEB-1992 14:00:57.94
FINDPATTERNS.FOR;1   11-MAR-1991 11:47:12.53 

Here is a sample of the output... I can send the whole 1000+block file to anyone
who wants.  It just appears that my version of findpatterns works.  I can't 
explain it though... does anyone know what the actual bug is???

! FINDPATTERNS on Owl:* allowing 0 mismatches

!        1 <~V{1,49}~V>                              May 26, 1994  14:09 ..

          3dhq_Strco  ck: 5502  len: 34    ! 3-DEHYDROQUINATE DEHYDRATASE (EC 4.

1                     <~V{1,49}~V>

          48kd_Bacce  ck: 9418  len: 15    ! 48 KD PROTEIN (FRAGMENT). - BACILLU

1                     <~V{1,49}~V>
             1:       XTQQEGMDISSSLXK



Keith O. Elliston                       |  Phone: (908)594-6099                 
Dept. of Bioinformatics                 |    Fax: (908)594-2929  
Merck Research Laboratories             |  Email: Elliston at merck.com 
Box 2000 / Mail Stop RY80A-1            |         Elliston at aol.com
Rahway, NJ 07065  U.S.A.                |         Elliston at mbcl.rutgers.edu

More information about the Info-gcg mailing list

Send comments to us at biosci-help [At] net.bio.net