AMBER Archive (2004)

Subject: Re: AMBER: why the enegy of native state of protein is high in GBSA calculation?

From: Carlos Simmerling (
Date: Fri Jun 04 2004 - 06:52:40 CDT

you need to read the literature more. it is
now generally accepted that ff99 has serious
bias and you should not expect the native state
of a protein with significant beta-sheet content
to be lowest in energy. try one of the more recent
force fields provided in AMBER 8.

Okur, A., Strockbine, B., Hornak, V. and Simmerling, C., ˇ®Using PC Clusters
to Evaluate the Transferability of Molecular Mechanics Force Fields for
Proteinsˇ±, J. Comput. Chem., 24:21, 2003.

----- Original Message -----
From: "J. Zhang, Dr" <>
To: <>
Sent: Friday, June 04, 2004 5:21 AM
Subject: AMBER: why the enegy of native state of protein is high in GBSA

> Dear Amber users,
> Sorry for that this letter is relatively long.
> I encountered a strange problem in my GBSA run. I found
> the potential and total energy of the native state
> of protein CI2 is even higher than that of non-native
> state. But this is not correct.
> I made two runs at temperature 100K, one was started from
> the NMR structure of the native state of protein CI2, the
> other was started from a relative random structure (see
> the attached pdb files). Minimizations have been made before
> the MD runs. In the calculations, the ff99 force field
> and AMBER7 is used.
> The mdin files in two runs are the same:
> ------------------------------------------
> constant temperature GBSA run
> &cntrl
> imin=0, ntb=0, ntt=1, ntc=2, ntf=2, dt=0.002, cut=8.0,
> igb=2, gbsa=1,
> ntpr = 10, ntwr=10, ntwx=10, ntwe=10,
> nstlim = 1000000,
> temp0 = 100,
> tempi = 100,
> tautp = 0.1,
> &end
> ------------------------------------------
> The obtained energies (from mden files) are given in
> the attached JPG file. To compare the two trajectories,
> I put the energies of two trajectories together in
> each figure. Long time simulations have been made to
> ensure equilibrium, but here I only show the results
> of a short time.
> I can be seen that the potential and total energy of
> native state are even higher that that of non-native state!
> But this is not correct. The starting structures of two
> runs are compared to the finial structures, and they are
> almost the same. That is, the structures did not change much.
> Similar results have also been seen in temperature 300K.
> Why? Is this due to some incorrect parameters in my mdin file,
> or due to improper force field chosen? or sth else?
> Thank you very much for your help.
> Best Regards,
> ---
> J. Zhang, Dr,
> Institute of Biophysics,
> Nanjing University

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to