Hi Pengfei and Chris
Thanks for your inputs. It does clear few things.
I tried a bonded model but it did not go well with the water molecules
(i think you already foresaw this in your tutorial). So i built a non
bonded model for all 6 bonds. But i sure can make a hybrid model with water
molecules as non bonded and others bonded. From your experience what will
be the profit of using a hybrid model.
I checked the RCSB IDs 2CNQ and 1KJQ and you are right, they do show only
2 ADPoxygen-Mg2+ contacts in Mg2+'s first shell. Given that these are 1
Angstrom structures we should consider it as relevant.
I also noticed the both seem to have 6 bonds with Mg2+ in octahedral
form (for 2CNQ : 4 water 2 ADPoxygen; for 1KJQ: 2 water 2 ADPoxygen; 1
oxygen from GLU267 and 1 oxygen from GLU279 ).
With this information i would say best guess from my side would be to go
with 2CNQ kind of structure of 4 water 2 ADPoxygen. As i want to study the
dissociation of ADP from the protein.
Crystallizing the protein with ADP is out of question as many researchers
have already tried and it just does not work out. While there is
literature, none says anything about the ADP water conformation. Its just
that ADP is formed and we see conformational change. I think understanding
the causes of this conformational change is needed.
Hence i need to model a ADP in place of ATP along with Mg2+ and its
co-ordinate bonds.
What do you suggest will be a reasonable starting structure in such a
case.
==================================================================================================================
To put some light on my work here:
In my case i have 2 PDBs to work with:
1. an electron microscopy PDB of little over 3 Angstrom resolution of the
protein with ATP. It does not have water molecules. (All protein domains
available)
2. a X-ray structure of ATP binding domain of the protein with ATP with 3
ATP-oxygen bond, 2 water, and 1 Oxygen atom from amino acid. (only ATP
binding domain was crystallized in this case)
We know that ATP is hydrolyzed to ADP and then conformation changes. I am
interested in understanding the confrontational changes after ATP->ADP
hydrolysis, hence i need to model ADP with MG and water. A full Quantum run
is not possible for nor a QM/MM as it might be required to run over 100s of
nanoseconds.
====================================================================================================================
On Sun, Jun 3, 2018 at 4:35 AM Chris Neale <candrewn.gmail.com> wrote:
> Mg2+/GDP does not always have three ADP oxygens in the first shell of Mg2+.
> See, for instance, the two highest resolution structures in the PDB with
> ADP and Mg2+ (PDBs: 2CNQ and 1KJQ), which each have only 2 direct
> ADPoxygen-Mg2+ contacts. Given that the lifetimes of atoms in Mg2+'s first
> shell are likely much longer than what you can achieve via simulation, I'd
> try to find some experimental evidence for the Mg2+ coordination in your
> particular system.
>
> On Sat, Jun 2, 2018 at 4:07 PM, Pengfei Li <ambermailpengfei.gmail.com>
> wrote:
>
> > Hi Abhilash,
> >
> > Do you want to model all the 6 coordination bonds with the bonded model?
> >
> > -Pengfei
> >
> > > On May 31, 2018, at 12:45 PM, Abhilash J <md.scfbio.gmail.com> wrote:
> > >
> > > Hi everyone,
> > >
> > > I plan on simulating a ADP with one MG2+ion and protein.
> > > Will it be plausible to take a Mg2+ with 6 co-ordinate bonds. Three
> > with
> > > Oxygen atoms of ADP (in the phosphates) and three with oxygen atoms in
> > > water molecules (Octahedral form).
> > > I can parameterize this using MCPB.py or Gaussian to get a bonded or
> a
> > > non bonded model to run simulation in it.
> > > I don't want to keep bonds with the protein as i need to study the
> > > dissociation of ADP from the protein.
> > > Is the ADP + MG2 + 3 water in Octahedral form configuration a
> > reasonable
> > > assumption for a starting structure.
> > >
> > >
> > > Regards
> > >
> > > Abhilash
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 05 2018 - 12:00:03 PDT
