Hi Charles,
Thanks a lot for your reply. I didn't analyze the radius of gyration of
"*_ave.pdb", but all the others. Calculations were performed 100 times and
50 with lowest energies were analyzed. That is, with Ne = 1, 50 pdb files,
with Ne = 8, 400 pdb files were analyzed. The Rg of 50 (or 400) structures
were averaged. I still don't get why the results without restraints are
different in different size of ensemble......
Is that possible to use XplorPot ("BOND",...) instead of AvePot(XplorPot,
"BOND",...) in ensemble calculation (for Ne > 1), if I don't want these
terms to be averaged across the ensemble members?
Regards,
Jie-rong
2009/2/24 <Charles at schwieters.org>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> Hi Jie-rong--
>
> >
> > The attachments are the input files that I used.
>
> The mailing list doesn't allow large emails- these messages are sent to
> many people.
>
> >
> > > I used a modified script from dna_refi/ensemble.py, in which only BOND,
> > ANGL,
> > > IMPR, VDW energies are considered, to test how the random conformers
> look
> > > like.
> > >
> > > The size of ensemble for 1 and 8 were tested. As you suggested before,
> the
> > > larger the size of ensemble, the longer dynamics needed. However, even
> if
> > > the dynamics for 8 conformer is set up to 12 ps (at each cooling
> > > temperature, default in the original script was 0.2 ps), the radius
> of
> > > gyration of 8 conformer is much smaller than the results from 1
> conformer.
> > I
> > > would expect the outcome of calculation without experimental restraint
> > should
> > > be independent of the size of ensemble. (It doesn't change this
> phenomenon
> > > if RAMA is added. )
> > >
> >
>
> I ran the script and generated 4 ensembles of Ne=8 structures. The
> average Rg for these was 26.7 +/- 4.0 I then ran two calculations of
> 4 Ne=1 structures and got Rg= 36.0 +/- 4.3 and 30.0 +/- 3.2
>
> There may be something systematic going on, but it's not too much. Note
> that this calculation starts with an extended structure with extremely
> short bonds- not at all near equilibrium. It might be interesting to
> watch trajectories of these runs.
>
> Note also that the Ne=8 average structures are meaningless.
>
> best regards--
> Charles
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>
>
> iD8DBQFJpHH8PK2zrJwS/lYRAqXvAJ4lnCtdz1e2UgpQBvfBZ8zpJ/x3uQCfX304
> MJwvfe9FTbEg27+l55050gw=
> =8MIV
> -----END PGP SIGNATURE-----
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20090224/56ab900e/attachment.html
