AMBER Archive (2008)

Subject: Re: AMBER: High force constant for MD ???

From: Waqas Nasir (nasirwaqas1983_at_yahoo.com)
Date: Thu Jul 24 2008 - 05:42:13 CDT

Hi,

Thanks Carlos for a quick reply. Well, I agree that there is somethig fishy going on. I have run the MD equilibration with 50 and 200 KCal/mol force constants on fucose ring and both of them have distorted the other 3 main rings in tetrasaccharide that I have in the complex with protein. The parameter file that I am using is Glycam_06d.dat on http://www.glycam.com/CCRC/biombuilder/biomb_index.jsp

When I build the tetra saccharide there are a couple of bonds that are missing so I build them up mannually by drawing them through xleap. When I save the parameter and topology files then xleap displays several warning messages. Following is the excerpt from that text,

----------------------------------------------------------------------------------------------------------------------------------
 ** Warning: No sp2 improper torsion term for  H1-N-CG-CG
        atoms are: H2 N2 C2 C1
 ** Warning: No sp2 improper torsion term for  C-HC-CG-HC
        atoms are: C2N H1M CME H2M
 ** Warning: No sp2 improper torsion term for  C-HC-CG-HC
        atoms are: C2N H1M CME H3M
 ** Warning: No sp2 improper torsion term for  C-HC-CG-HC
        atoms are: C2N H2M CME H3M
 ** Warning: No sp2 improper torsion term for  HC-HC-CG-HC
        atoms are: H1M H2M CME H3M
old PREP-specified impropers:
 <3YB 839>:  C2   C2N  N2   H2N
 <3YB 839>:  CME  N2   C2N  O2N
 <0vA 842>:  C2   C2N  N2   H2N
 <0vA 842>:  CME  N2   C2N  O2N
 total 1972 improper torsions applied
 4 improper torsions in old prep form
Building H-Bond parameters.
Not Marking per-residue atom chain types.
Marking per-residue atom chain types.
  (Residues lacking connect0/connect1 -
   these don't have chain types marked:

    res    total affected

    CGLY    2
    NTHR    2
  )
 (no restraints)
----------------------------------------------------------------------------------------------------------------------------------

I ignored these warnings in the beginning but I dont know now if they are hinting towards any possible problem. I would appreciate if you could help me out!

Thanks a lot for the help,

Stay blessed!
Kind regards,
Waqas.

----- Original Message ----
From: Carlos Simmerling <carlos.simmerling_at_gmail.com>
To: amber_at_scripps.edu
Sent: Wednesday, July 23, 2008 2:02:13 PM
Subject: Re: AMBER: High force constant for MD ???

yes that is very high and may cause problems.
maybe it is better to determine why you need such a strong force
constant- why does the sugar have such large desire to become distorted?
it sounds like you have a problem in the parameters, and if it takes
700 kcal/mol force constant to fix it then things are really wrong.
usually the underlying parameters are several kcal, not hundreds.
I think for us to help more we need to know more about your
parameters for the carbohydrate and also what you mean by "valid"
conformation. my concern in reading a paper about such work is that
if you need 700 kcal to fix the part you know is "wrong", then probably
the part you are trying to learn from the simulation is also wrong.
carlos

On Wed, Jul 23, 2008 at 5:49 AM, Waqas Nasir <nasirwaqas1983_at_yahoo.com> wrote:
> Hi,
>
>
 Hope every thing is working fine at your side.
>
> Well, I am running minimization and MD studies using sander module of amber
> version 8 on protein carbohydrate complex. I have to fix the sugar ring so
> that the conformation remains valid till the end. In minimization, I first
> tried the force constant of 50 K-Cal/Mol-A**2 which didnt work that well and
> the ring was distorted, then I jumped to a higher value of 700 (just to make
> sure that the ring does not move!) which worked out in the end.
>
> Now I am attempting to do the equilibration. Is there any problem if I use
> the same force constant of 700 in equilibration and MD as well (just on the
> sugar ring not on the whole complex). I read in the tutorial that its not
> recommended to use high force constants in MD simulations. Any body with any
> comments or suggestions?
>
> Thanks alot for
 reading,
> Stay blessed!
>
> Regards,
> Waqas.
>
>
> PS: Thanks Lachele, for your extensive and worthed help!
>
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu