-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello Yi--
> Thank you for your reply. I did modify the 'newrefine.py'(attached file) but
> got the following errors:
>
> InternalDynamics::step: large timestep detected. Halving.
> %atoms "ALT2-35 -CYSP-SG2 " and "ALT3-35 -CYSP-SG2 " only 0.01 A apart
> InternalDynamics::step: large timestep detected. Halving.
> InternalDynamics::step: large timestep detected. Halving.
> InternalDynamics::step: large timestep detected. Halving.
Clearly, something bad is happening but I can't tell what from this
output. Could you please send the 20 lines of output before these
messages are printed?
>
> Also, I am not quite understand the clock atom and tauc part. How are the
> initial values (resid 777-783; taucapp=2.9,3.0) chosen?
>
In the example in eginput/pre the correlation times are optimized and
the tauc values are encoded using clock atoms. Those experiments/tags
which should share tauc use the same clockResid. Of course, you could
also just fix this value and then forget about the clockResid.
> I want to calculate form extended structures, and the modified 'anneal.py'
> (attached file) did work. But in the output file, there are VDW violations
> between conformers of the same cysp like these:
>
> viol% amount index restraint name
> ---------------------------------------------------------------------------
> 100.0 0.96 0 ( ALT1 7 CYSP SG1 ) ( ALT2 7 CYSP SG2 )
> 100.0 1.01 1 ( ALT1 7 CYSP SG1 ) ( ALT3 7 CYSP SG2 )
> 100.0 1.02 2 ( ALT1 7 CYSP SG2 ) ( ALT2 7 CYSP SG1 )
> 100.0 1.85 3 ( ALT1 7 CYSP SG2 ) ( ALT2 7 CYSP SG2 )
> 100.0 1.18 4 ( ALT1 7 CYSP SG2 ) ( ALT2 7 CYSP CL1 )
>
> Are these normal?
You can ignore these violations (which you certainly don't care
about). Unfortunately, it's currently not easy for me to suppress this
output. If you're working from newRefine.py, there should be a .bumps
file created which lists nonbonded violations which are consistent with
the current contraints interactions settings.
> I use MOLMOL to calculate the distance between 'OS1' and
> 'HN' atoms. The calculated ones are larger than the max distance in the input
> '.tbl' files, but they were not reported in the '.viols' files.
>
To get a list of all restraints in the .viols file (for verification),
add this line:
noe.setShowAllRestraints(True)
to the noe setup. If there truely is a discrepency between Xplor-NIH and
MOLMOL, it should be investigated.
best regards--
Charles
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.9 <http://mailcrypt.sourceforge.net/>
iEYEARECAAYFAkyHydIACgkQPK2zrJwS/lZD2gCeJkbNgZofl90GwyckImzqGKw+
aDgAnRMWHUQhkUn3kTRAuruQcv0KSQqX
=NaKB
-----END PGP SIGNATURE-----
