Hi Veronica,
the same input structure run with slightly different MD setups will produce
different trajectories especially for long simulations as tiny coordinate
differences accumulate and the system evolution is chaotic. You dont say
what you did to arrive at free energy results from your trajectory, but
different conformational ensembles will give different aggregate
properties. If you consider both your Amber18 and 23 trajectories realistic
ensembles, then there is no real result. Your two runs give you an
indication of how large the random error upon repeated calculation is going
to be (maybe too large to interpret the results in a meaningful way, which
is not at all uncommon for complex free energy calculations). Try to
understand the differences in your simulation trajectories and how they
influence convergence of the results, that could help interpreting the
values you receive.
Kind Regards,
Thomas
On Fri, Sep 22, 2023 at 11:13 AM VERONICA MARTIN HERNANDEZ via AMBER <
amber.ambermd.org> wrote:
> **Warning** The sender address (VERONICA MARTIN HERNANDEZ via AMBER ) can
> not be verified, sender email address could be spoofed. Please take care to
> proceed.
> Hi everyone
>
> I have been running a MD of a protein-ligand complex with AMBER 18 the las
> three months, but now I tried the same MD with the same scripts in AMBER 23
> and the results are different; the MD that I have runned in AMBER 18
> provides a -19 kcal/mol result and the MD I run in AMBER 23 provides a -31
> kcal/mol result.
>
> What should I do? What are the real results?
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
--
*Dr. Thomas Steinbrecher*
Principal Scientist CADD
Roche Pharma Research and Early Development
Roche Innovation Center Basel
F. Hoffmann-La Roche Ltd
Bldg. 092/3.92
Grenzacherstrasse 124
4070 Basel
Switzerland
Phone +41 61 682 1319
mailto: thomas.steinbrecher.roche.com
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Sep 22 2023 - 02:30:03 PDT
