[Xplor-nih] Ensemble calculation of denatured protein

[Jie-rong Huang]

Dear Charles,

Some error messages at the first begining of terminal window that I didn't
notice last time:
############
calcD_G: singular D matrix: { { nan } }
H matrix: { { nan, nan, nan, 0, 0, 0 } }
node level: 227
number of children: 0
atoms:
        1228 GLY 76 C : { nan, nan, nan }
        1229 GLY 76 OT1 : { nan, nan, nan }
        1230 GLY 76 OT2 : { nan, nan, nan }
############
Is this the second possibility you mentioned?

The pdb and psf files input are both:
############
ATOM   1229  C   GLY    76     127.034   0.640  -0.873  1.00  1.00
ATOM   1230  OT1 GLY    76     127.291  -0.478  -1.288  1.00  1.00
ATOM   1231  OT2 GLY    76     127.580   1.657  -1.268  1.00  1.00
############
############
    1229      76   GLY  C    C      0.140000       12.0110           0
    1230      76   GLY  OT1  OC    -0.570000       15.9990           0
    1231      76   GLY  OT2  OC    -0.570000       15.9990           0
############
The atom numbers are inconsistent with those indicated in error message, is
it the problem? Thanks!!

Best wishes,

Jie-rong

2008/4/21, Charles at schwieters.org <Charles at schwieters.org>:
>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> Hello Jie-rong--
>
> >
>
> > Thank you very much for your reply.  I am not sure if I understand you
> > correctly. Do you mean the final temperature of cooling should be room
> > temperature (e.g. 298K) instead of 25K in the script? And the high
> > temperature dynamics at 3000K and cooling starting from 3000K are not
> changed
> > in my case?
> >
>
>
> To start with, just leave the temperatures alone- let's see how large
> an ensemble is needed for convergence.
>
>
> > Why does the script start from a folded structure instead of extended
> one? Is
> > that because there are some "high temperature" steps which can randomly
> > expand the structure, and therefore it doesn't matter the starting
> structue
> > is folded or expanded?
> >
>
>
> A best initial guess was used in this case. Wildly divergent, expanded
> conformations were not sampled.
>
>
> > Since I only have PRE data, I commented out "NOE", "dihedral angle
> > restraints", "repulsive restraints for EDTA-DNA phosphate", "planerity
> > restraints", and "base-base positional database", and related parameters
> > (e.g.k_noe) as well.
>
>
> sounds good. You may want the RAMA term. Remove the setup in the .py
> file, and replace it with:
>
> #Rama torsion angle database
> #
> import protocol
> protocol.initRamaDatabase()
>
>
> >  Now I am struggled, final lines in the log file is:
>
>
> there may be two problems here.
>
>
> 1)
>
> > test.py(569):          s.run()
> > *--- Dynamics ---- step=      0 ---- time=         0 ---- delta_t=
> 0.002
> > --*
> > | E(kin)+E(poten)=     21110.576 E(kin)=      4449.966 temperature=
> > 3238.358 |
> > |   E(poten)= 16660.6104132       grad=           nan      ANGL=
>
>
> something's wrong with the gradient. Please remove potential terms one
> at a time to figure out which is causing the nan. I suspect a PRE term.
>
> 2)
>
>
>
> > InternalDynamics::step: caught assertion failure:
> >    calcD_G: singular D matrix. Bad topology?
>
>
> There may also be a topology problem caused by a bad setup of the
> IVM. Let's fix problem 1) first, though.
>
> best regards--
>
> Charles
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.6 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>
>
>
> iD8DBQFIDLPWPK2zrJwS/lYRAuJJAJ9EhzW8YCVDSRFAuVmLdPJ7+8gcmgCfY+4j
> W2pE2jp913L60FZOLcGsRGk=
> =5sPb
> -----END PGP SIGNATURE-----
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20080422/b5a1ad1d/attachment.html