Hi Franz,
One cause of this that I've encountered is dihedral angle restraints
that have force constants that are too large. The cdih force constant is
the product of an overall weight factor S ("scale") set in the annealing
script, and an energy constant C set in the dihedral angle restraint
file. If the product of S and C is accidentally large, sa.inp can "blow up".
Even with the internal variable engine, large dihedral angle restraint
force constants can cause problems. In that case, the time step will
keep halving until it is too small, and the run will abort.
Keith
John Kuszewski wrote:
>Hi Franz,
>
>This is a fairly common problem using the 'classic' xplor dynamics
>engine. It can be caused by
>several things:
>
>1. Bad covalent restraints (bonds, angles, etc), caused by bad
>parameter sets. If you're using the
>current protein.par and protein.top files, that's probably not the
>issue.
>
>2. Bad distance restraints (and to a much lesser extent, bad
>covalent restraints). In this context,
>'bad' could mean too tight a distance range, too high a force
>constant, or inconsistent with each other.
>
>3. Bad luck. Try running with a different random number seed.
>
>Of these, the most common problem is distance restraints that are too
>tight or inconsistent with each other.
>Particularly if your temperature explosions are occurring at the
>beginning of the cooling phase,
>because that's when full vdW restraints are added in, and so some
>distance restraints that could
>be satisfied by coordinates that included vdW overlap suddenly start
>fighting with the vdW term.
>
>N.B. If you're using Charles's internal variable dynamics engine, you
>can set it to avoid this problem
>a lot of the time by allowing it to take variable-sized timesteps.
>That prevents the simulation from
>injecting extra energy into the system in regions where the potential
>energy surface is changing
>steeply.
>
>Hope this helps.
>
>JK
>
>On Nov 27, 2007, at 1:48 PM, Franz Hagn wrote:
>
>
>
>>Hi all,
>>
>>during the xplor calculation (with sa.inp script and xplor 2.16),
>>after
>>high temperature dynamics, I get unusual high temperature and abortion
>>of the run. Does anybody know some hints? Are my restraints to
>>rigorous ?
>>Thanks, Franz
>>_______________________________________________
>>Xplor-nih mailing list
>>Xplor-nih at nmr.cit.nih.gov
>>http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
>>
>>
>
>_______________________________________________
>Xplor-nih mailing list
>Xplor-nih at nmr.cit.nih.gov
>http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20071128/f7a6154f/attachment.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: keith.constantine.vcf
Type: text/x-vcard
Size: 453 bytes
Desc: not available
Url :
http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20071128/f7a6154f/attachment.vcf
