Mary,
Very likely it is due to the lack of convergence. Assuming that your
MD behaves well, i.e. not unfolding the protein etc. 10ns seems very
short for multi-traj MMPBSA of typical-sized proteins.
A straightforward but slow way to check convergence is to restart your
MD for 10ns each time and apply MMPBSA with snapshots from each of the
following sets of trajectories:
10ns, 20ns, 30ns, ... 100ns and so on.
You'll get a MMPBSA average energy for each of the above sets. Plot
these cumulative average energies versus the simulation times (10ns,
20ns, ..., 100ns, and so on) and you'll see whether the cumulative
average is converged, i.e. not changing much any more.
Of course if you have a good statistical software, you may be able to
get a plot of the cumulative average over simulation time given all
the snapshots at once (for example all of the 100ns).
All the best,
Ray
--
Ray Luo, Ph.D.
Professor
Biochemistry, Molecular Biophysics, Chemical Physics,
Chemical and Biomedical Engineering
University of California, Irvine, CA 92697-3900
On Mon, Mar 21, 2016 at 7:06 PM, Mary Varughese <maryvj1985.gmail.com> wrote:
> Sir,
>
> Infact i have tried multiple trajectory approach also (10 ns trajectory).
> But didnt get any favorable result.
>
> May be because i havent considered "make sure the average
> delta G has converged by checking cumulative averages of all three
> runs (complex, receptor, and ligand)". Would you please explain cumulative
> average?
> And also about entropy estimation. ? It would be really helpful.
>
> Thanking you for your time and reply
>
> mary
>
>
> On Tue, Mar 22, 2016 at 6:09 AM, Ray Luo <rluo.uci.edu> wrote:
>
>> Hi Mary,
>>
>> If you have a flexible ligand, the single trajectory approach is
>> probably not the best way to go. Please try the multi-trajectory
>> approach. Apparently, a key point here is to make sure the average
>> delta G has converged by checking cumulative averages of all three
>> runs (complex, receptor, and ligand). However, some sort of entropy
>> estimation is also important to take into account the conformational
>> flexibility in the delta G calculation.
>>
>> All the best,
>> Ray
>> --
>> Ray Luo, Ph.D.
>> Professor
>> Biochemistry, Molecular Biophysics, Chemical Physics,
>> Chemical and Biomedical Engineering
>> University of California, Irvine, CA 92697-3900
>>
>>
>> On Fri, Mar 18, 2016 at 9:48 PM, Mary Varughese <maryvj1985.gmail.com>
>> wrote:
>> > sir,
>> >
>> > these are the files used to calculate PBTOT and entropy. I have done it
>> > with other ligands(more rigid ligands) successfully. The problem here is
>> > that the current ligand is half part flexible(a ch2-ch2-ch2-ch3 flexible
>> > chain). Though the ligand bind experimentally and
>> theoretically(throughout
>> > the simulation time) the movement of the flexible region is causing
>> changes
>> > in PBTOT and entropy such that i cant get a statistically reliable value
>> > (the final value obtained are highly positive; when i check the values
>> (BE)
>> > for each frame its deviating very much). The flexible part is causing
>> that.
>> >
>> > So i would like to know which strategy i should be adopted in such cases.
>> >
>> >
>> > thanking you for ur reply
>> >
>> > On Sat, Mar 19, 2016 at 6:43 AM, Ray Luo <rluo.uci.edu> wrote:
>> >
>> >> Mary,
>> >>
>> >> Maybe a bit more info is helpful, i.e. your inpu file. Also please note
>> >> that mmpbsa single trajectory approach is more useful for delta delta G
>> >> estimation.
>> >>
>> >> All the best,
>> >> Ray
>> >> On Mar 16, 2016 5:20 PM, "Mary Varughese" <maryvj1985.gmail.com> wrote:
>> >>
>> >> > Sir,
>> >> >
>> >> > I run some DNA-ligand1 complexes. Though the ligand remains bind
>> >> throughout
>> >> > the simulation, (the ligand binds experimentally also), I am not
>> getting
>> >> a
>> >> > favorable binding energy from MMPBSA (single trajectory approach). The
>> >> > ligand has a CH2-CH2-CH2 chain on one end which causes the ligand some
>> >> > movement (entropy changes) and hence a stable binding energy is not
>> >> > possible. Is there any other way to quantify the binding energy. Would
>> >> you
>> >> > please suggest a reliable approach in such situations where the ligand
>> >> has
>> >> > much flexibility?
>> >> >
>> >> > thanking you
>> >> > mary
>> >> > _______________________________________________
>> >> > 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
>>
> _______________________________________________
> 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 Mon Mar 21 2016 - 23:30:04 PDT