AMBER Archive (2005)

Subject: AMBER: GBSA simulation problem

From: Guillaume Bollot (Guillaume.Bollot_at_chiorg.unige.ch)
Date: Mon Oct 31 2005 - 02:23:26 CST


Dear all Users!
I would like to perform molecular dynamic on a protein pore (3600 atoms) in
continuum with Amber 8.0.
My problem is with this system, the simulation seems to be very very long
(md1: 57 hours for 100ps). Is it normal?
Amber was compiled on AMD64 under Linux (Gentoo) and my input files are :

--------------------------------------------------
min1.in:
&cntrl
imin=1, maxcyc=1000, ncyc=500,
drms=0.05, igb=1,
ntb=0,
cut=12.0,
ntpr=100,
&end
-------------------------------------------------
md1.in
&cntrl
   imin = 0,
   irest = 0,
   ntb = 0,
   cut = 12, igb=1,
   ntr = 1,
   tempi = 300.0,
   temp0 = 300.0,
   ntt = 1,
   nstlim = 100000, dt = 0.001,
   ntpr = 100, ntwx = 100, ntwr = 100,
&end
Restraint on Pore (1-256)
0.5
RES 1 256
END
END
---------------------------------------------------
md2.in:
&cntrl
   imin = 0,
   irest = 1, ntx=5,
   ntb = 0,
   cut = 12, igb=1,
   ntr = 0,
   tempi = 300.0,
   temp0 = 300.0,
   ntt = 1,
   nstlim = 400000, dt = 0.001,
   ntpr = 400, ntwx = 400, ntwr = 400,
&end
------------------------------------------------------
I would like to know if the simulation time is normal and my methodology is
correct?
I performed already the Tutorial 2 to estimate the binding energy and my
simulation were ok.
Thanks in advance!!!!

BOLLOT Guillaume
Organic Chemistry Department
University of Geneva
30, quai Ernest-Ansermet
CH-1211 Geneva 4
tel. +41-22 379 6155
e-mail: Guillaume.Bollot_at_chiorg.unige.ch

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu