I can't open excel attachments. send only text or jpg.
did you calculate the rmsd of the protein only? is it fine?
did you start from a known structure? or did you have to model the peptide?
your equilibration is very simplistic. usually a much more slow,
multi-step procedure is used in which the solvent and hydrogen and
anything else not determined by experiment is equilibrated and the
rest kept restrained. then, these restraints are slowly turned off.
too-rapid equilibration can lead to instability.
On Fri, Aug 5, 2011 at 5:36 AM, Xiao Chen <chen.xiao.po.gmail.com> wrote:
> Dear Carlos,
>
> Thanks for the email.
>
> The protein-peptide complex which I am trying to simulate has 225 residues
> and the peptide is 10 residues long. The peptide interacts with the protein
> through 3 hydrogen bonds at one end. The residues at the other end of the
> peptide are free (non interacting).
>
> First, I ran 2000 cycles of minimization on the complex after creating the
> .prmtop file form tleap.
>
> Next, I equilibrated the complex at constant volume, temperature and then
> pressure.
>
> Then i collected production run for 15ns (1500 frames)
>
> After the run, I calculated the r.m.s.d. of coordinates generated using
> ptraj:
>
> trajin abc.mdcrd 1 1500 1
> reference ref.pdb
> rms reference out rmsd.dat .CA, C, N, O
> go
>
> r.m.s.d of the coordinates generated during the simulation, fluctuate a
> great deal and then jump sharply after ~10ns and progressively keeps
> rising.
> Find attached the r.m.s.d. graph over 15ns (1500 frames).
>
> I looked at the trajectory. Over these 15ns the peptide fluctuates such that
> at one point (after about 10ns) the distance between the CA of a peptide
> residue (which makes hydrogen bond with the protein) is ~10 A with the
> reference. Please help!
>
> Thanks,
>
> Po
>
>
>
> On Thu, Aug 4, 2011 at 4:44 PM, Carlos Simmerling <
> carlos.simmerling.gmail.com> wrote:
>
>> You don't tell us very much. For example which atoms are used in rmsd? It
>> is
>> the part that you said is expected to fluctuate? Also, what did you do to
>> calculate rmsd? Did you use any sort of periodic imaging? Have you
>> visualized this trajectory to see what is happening? My rule is to always
>> look at your primary data (coordinates, through a viz program) and the do
>> numerical analysis to quantify what you see in the data. Never, ever skip
>> the step of actually looking at the molecule.
>> On Aug 4, 2011 3:00 AM, "Xiao Chen" <chen.xiao.po.gmail.com> wrote:
>> > Dear All,
>> >
>> > I am trying to simulate a protein-peptide complex. the peptide is ~10
>> > residues out of which 3 at one end are interacting with the protein. The
>> > other end of the peptide is free to fluctuate. I ran a simulation of 10ns
>> > and it was observed that the r.m.s.d of the coordinates generated during
>> the
>> > simulation becomes very high after ~2ns and progressively keeps rising
>> after
>> > that. This could indicate that something is not okay with the simulation.
>> > Can you suggest what I can alter to stabilize the simulation?
>> >
>> > Thanks,
>> >
>> > Po
>> > _______________________________________________
>> > 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 Fri Aug 05 2011 - 03:30:04 PDT
