VMD-L Mailing List

From: Vlad Cojocaru (vlad.cojocaru_at_mpi-muenster.mpg.de)
Date: Wed Nov 30 2011 - 11:15:15 CST

Hi John,

Thanks a lot for this answer. I contacted the stride developers and they
seem to be aware that the compilation of the stride code on new linux
distributions leads to non-functional executable.
They will get back to me regarding the issue. The limits in stride.h do
not influence this ... I compiled previously with these limits (500) and
everything worked fine ...

In the meantime I am using the stride_LINUXAMD64 that comes with vmd 1.9
but this is not compiled with the increased limits ... As I need those
limits increased (200 should work just fine) for some systems, maybe it
would be nice if you could send me a new executable. You can also send
me the modified source but I think this will have the same problem on my
OS... But maybe for a trial wiould be nice

Thanks
Vlad

On 11/30/2011 05:24 PM, John Stone wrote:
> Vlad,
> Sorry for the delayed reply, I'm catching up on emails after being
> out of the office for a few days. We don't really do anything special
> when compiling the STRIDE binary that is provided along with VMD from
> source code. The compilation is done using typical GCC flags for each
> target platform.
>
> Perhaps your changes to increase the residue atom count values to
> 500 broke something inside STRIDE?
>
> We have modified our copy of the STRIDE source code to allow more atoms
> in each residue, and to cut out some computations that
> VMD doesn't want or need, and we've inserted a call to one of VMD's
> fast gridsearch routines to speed up the code that finds the H-bonds.
>
> The two most generic changes we made for our version are shown below:
>
> % diff stride.h ../src.orig/stride.h
> 45,46c45
> < /* #define MAX_AT_IN_RES 75 */
> < #define MAX_AT_IN_RES 200
> ---
>> #define MAX_AT_IN_RES 75
> % diff stride.c ../src.orig/stride.c
> 96c96
> < return(SUCCESS);
> ---
>> return(0);
> 286c286
> < Cmd->ExposedArea = NO;
> ---
>> Cmd->ExposedArea = YES;
> If you like, I can send you our modified STRIDE source so you can
> play with it.
>
> Cheers,
> John Stone
> vmd_at_ks.uiuc.edu
>
> On Thu, Nov 24, 2011 at 03:50:40PM +0100, Vlad Cojocaru wrote:
>> Dear VMD developers and users,
>>
>> I am having a strange problem with using the stride executable with VMD
>> after an upgrade of the operating system from openSUSE 11.4 to openSUSE
>> 12.1
>>
>> I always used a customized stride executable that I compiled myself
>> because the default limits in stride.h are too stringent for my systems...
>> After the OS upgrade, I got errors (see below) every time I was trying
>> to visualize secondary structure. Before the upgrade everything worked
>> properly
>>
>> Looking in more detail, I noticed that my stride executable is not
>> producing any output when run on any PDB file. For comparison, the
>> executable provided with VMD 1.9 is working correctly. I tried
>> recompiling my stride with different versions of GCC (3.3, 4.3, 4.6) but
>> although the compilations ran smoothly with no error, the resulting
>> executable was still not functional.
>>
>> Has anyone seen this problem ? How was the stride_LINUXAMD64 binary
>> provided with VMD 1.9 compiled ? Any tricks ? I would need a stride
>> executable compiled with the line 43, 44 in stride.h modified as follows:
>>
>> #define MAX_AT_IN_RES 500
>> #define MAX_AT_IN_HETERORES 500
>>
>> Thank you for any suggestions
>>
>> Best wishes
>> Vlad
>>
>>
>> My arch is x86_64 (AMD)
>> OS: openSUSE 12.1
>> VMD: 1.9 and various CVS versions, all compiled by myself (GCC 4.6) ...
>> (the problem is though the same is I use the distributed VMD 1.9
>> binaries with my stride executable)
>>
>> ~~~~ - 0 ( 0 ) 215 ( 215 ) 3.5
>> ERROR) Unable to find Stride output file: /tmp/filefxWN5v
>> ERROR) Stride::read_stride_record: unable to read output file from Stride
>> ERROR) Call to Stride program failed.
>>
>>
>>
>> --
>> Dr. Vlad Cojocaru
>> Max Planck Institute for Molecular Biomedicine
>> Department of Cellular and Developmental Biology
>> Roentgenstrasse 20
>> 48149 Muenster, Germany
>> tel: +49-251-70365-324
>> fax: +49-251-70365-399
>> email: vlad.cojocaru[at]mpi-muenster.mpg.de
>> 

-- 
Dr. Vlad Cojocaru
Max Planck Institute for Molecular Biomedicine
Department of Cellular and Developmental Biology
Roentgenstrasse 20
48149 Muenster, Germany
tel: +49-251-70365-324
fax: +49-251-70365-399
email: vlad.cojocaru[at]mpi-muenster.mpg.de