Re: [AMBER] Magnesium charges using leaprc.ff99SB - metals move away

From: Mallipeddi, Premalatha <Premalatha.Mallipeddi.STJUDE.ORG>
Date: Fri, 9 Aug 2013 10:40:52 -0500

Dear Dr. Case,

          Thank you for your response. I see the metals start to move away from each other. The initial distance between the two magnesium atoms is 3.5 A. By 1 ns the distance is 6 A. Also, there are three oxygens of ligand coordinating to the two magnesium in the crystal structure, in addition to binding site residues coordinating the magnesium ions. This coordination geometry that is required for bimetal chelation is disturbed and the key interactions oxygens in ligand started to move away from the metal.

What do you suggest?

I tried the desc command, it works for protein residues but not for MG.

The pdb reads as

ATOM 3209 OXT ARG 196 -0.163 20.427 -10.286 1.00 0.00
TER
ATOM 3210 MG MG2 197 -4.194 5.140 -1.953 1.00 0.00
TER
ATOM 3211 MG MG2 198 -5.935 2.460 -4.047 1.00 0.00
TER
ATOM 3212 Cl UNK 199 -2.794 6.344 -12.114 1.00 0.00
And so on

> desc MET.1.N
ATOM
             Normal Perturbed
Name: N N
Type: N N
Charge: -0.4157 0.000
Polarization: 0.0000 0.000
Element: N (not affected by pert)
Atom flags: (decimal 131072 hex 0x20000)
        posfxd n posblt n posdrwn n selected n
        pert n notdisp n touched n posknwn Y
        internal n needsmin n needsbuild n
Atom position: 3.325770, 1.547909, -0.000002
Atom velocity: 0.000000, 0.000000, 0.000000
  Bonded to .R<MET 1>.A<H 2> by a single bond.
  Bonded to .R<MET 1>.A<CA 3> by a single bond.
> desc MG2.197.MG
STRING (with no reference): 'MG2.197.MG'

The command to load force field is: - bsub -P Pan -app amber12-ib -Is -q interactive tleap -f leaprc.ff99SB
I am using the following version to submit jobs in cluster: module load amber/12/openmpi-1.6.0-intel

Thanks and regards,
Latha.


________________________________
Email Disclaimer: www.stjude.org/emaildisclaimer
Consultation Disclaimer: www.stjude.org/consultationdisclaimer
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 09 2013 - 09:00:04 PDT
Custom Search