Hi,
The RMSD plot: https://imgur.com/a/WbUAPhI <https://imgur.com/a/WbUAPhI>
And the 2D RMSD plot: https://imgur.com/a/AZyu0jd <https://imgur.com/a/AZyu0jd>
As you can see, the ligand shows stable values (~0.1-0.6). In the case of the protein, the RMSD is higher, but the overall results seem fine; even the 2D RMSD looks good. The only "instability" appears to be in the membrane to which the protein is anchored.
I don't see any reason to assume there's an issue with the system's equilibrium. Rather, the issue might lie in how the MMPBSA topologies are stripped and the masks used. Perhaps switching to gb=2 instead of gb=0 could also be beneficial?
Best
–
Maciej Spiegel, MPharm PhD
assistant professor
.GitHub <
https://farmaceut.github.io/>
Department of Organic Chemistry and Pharmaceutical Technology,
Faculty of Pharmacy, Wroclaw Medical University
Borowska 211A, 50-556 Wroclaw, Poland
> Wiadomość napisana przez 王世玉 via AMBER <amber.ambermd.org> w dniu 11.09.2024, o godz. 03:36:
>
> Thank you for your mentioning the paper (10.3389/fphar.2022.1018351) . I am the author of this paper.
> Can you present your RMSD plot in your email.
>
> From my prespective, 100ns is too long, it hard for a liagnd binding to protein stably during the whole molecular dynamics. So RMSD (2D RMSD is better) is required to select while frames should be used to calculate the MMPBSA.
>
>
>
> > -----原始邮件-----
> > 发件人: "Maciej Spiegel via AMBER" <amber.ambermd.org>
> > 发送时间: 2024-09-11 03:17:15 (星期三)
> > 收件人: "Carlos Simmerling" <carlos.simmerling.gmail.com>
> > 抄送: "AMBER Mailing List" <amber.ambermd.org>
> > 主题: Re: [AMBER] Positive value of MM-GBSA computations
> >
> > > Implicit solvent can be challenging for membrane bound systems. Read the literature for similar work and see what they say about expected accuracy, especially when the results were validated against experiments.
> >
> >
> > I suppose so! After running MMPBSA, all binding energies are positive. And I don’t mean "close to 0," but at least 6 kcal/mol or more. I wonder if the issue stems from the membrane-bound system, just like you mentioned. Unfortunately, I can't find any paper that clarifies this issue; the docking score was negative, and according to this paper (10.3389/fphar.2022.1018351), the output of MMGBSA or MMPBSA should be consistent with docking.
> >
> > The system is well-equilibrated, as I ran it for 100 ns, and the RMSD values for the membrane and protein are fine.
> >
> > Also, skimming over the trajectory frames, nothing unusual happens, and the ligand remains in the binding pocket. If ΔG > 0, wouldn’t it just leave?
> >
> > I suspect the issue might be with the MMPBSA input file, which in my case is very simple. Maybe changing the solvent model (e.g., to igb=2) or using the entropy rotor correction (entropy=1) would help? Any ideas? This has already taken me two days. Thanks!
> >
> >
> >
> > –
> > Maciej Spiegel, MPharm PhD
> > assistant professor
> > .GitHub <https: farmaceut.github.io=""></https:>
> >
> > Department of Organic Chemistry and Pharmaceutical Technology,
> > Faculty of Pharmacy, Wroclaw Medical University
> > Borowska 211A, 50-556 Wroclaw, Poland
> >
> > > Wiadomość napisana przez Carlos Simmerling <carlos.simmerling.gmail.com> w dniu 08.09.2024, o godz. 19:37:
> > >
> > > A few thoughts
> > > The net binding energy is essentially zero.i would not consider the sign for such a small number. Perhaps this us a reasonable answer for this ligand and pose.
> > >
> > > Implicit solvent can be challenging for membrane bound systems. Read the literature for similar work and see what they say about expected accuracy, especially when the results were validated against experiments.
> > >
> > > Concerning gb vs pb, on my opinion gb is a poor choice for such calculations. It is highly approximate, often not trained for ligand chemical species, and is most useful when speed is crucial. This is the case for MD, but not for postprocessing.
> > >
> > >
> > > On Sun, Sep 8, 2024, 1:27 PM Maciej Spiegel via AMBER <amber.ambermd.org <mailto:amber.ambermd.org="">> wrote:
> > > Dear Amber Users,
> > >
> > > I am currently working on a protein anchored in a membrane bilayer. After equilibrating the system for 100 ns, I docked the potential ligand using AutoDock Vina and ran another 100 ns of simulations. The ligand consistently occupied the binding site. Subsequently, I performed MM-GBSA calculations and obtained a small but positive value for the interaction energy:
> > > -------------------------------------------------------------------------------
> > > -------------------------------------------------------------------------------
> > > Differences (Complex - Receptor - Ligand):
> > > Energy Component Average Std. Dev. Std. Err. of Mean
> > > -------------------------------------------------------------------------------
> > > VDWAALS -11.9224 2.2582 0.0226
> > > EEL -60.8192 15.3864 0.1539
> > > EGB 75.4142 12.6881 0.1269
> > > ESURF -2.6166 0.1865 0.0019
> > >
> > > DELTA G gas -72.7415 14.6457 0.1465
> > > DELTA G solv 72.7976 12.6124 0.1261
> > >
> > > DELTA TOTAL 0.0561 3.1697 0.0317
> > > -------------------------------------------------------------------------------
> > > -------------------------------------------------------------------------------
> > >
> > > For your reference, the input used is as follows:
> > >
> > > -------------------------------------------------------------------------------
> > > &general
> > > strip_mask = :WAT:K+:Cl-:PA:PE:OL:PC",
> > > /
> > > &gb
> > > saltcon=0.150,
> > > /
> > > -------------------------------------------------------------------------------
> > >
> > > I generated topologies using the command:
> > >
> > > -------------------------------------------------------------------------------
> > > ante-MMPBSA.py -p MD.prmtop -c COMPLEX.prmtop -r RECEPTOR.prmtop -l LIGAND.prmtop -s :WAT:K+:Cl-:PA:PE:OL:PC -n :FST
> > > -------------------------------------------------------------------------------
> > >
> > > Given the small positive interaction energy, I am wondering what could be the reason for this result. How should I interpret this value in the context of the expected binding affinity toward the pocket where the ligand was docked? Would doing MM-PBSA be a „solution” for the issue?
> > >
> > > Thank you for your insights.
> > > –
> > > Maciej Spiegel, MPharm PhD
> > > assistant professor
> > > .GitHub <https: farmaceut.github.io="" <https:=""></https:>>
> > >
> > > Department of Organic Chemistry and Pharmaceutical Technology,
> > > Faculty of Pharmacy, Wroclaw Medical University
> > > Borowska 211A, 50-556 Wroclaw, Poland <https: www.google.com="" maps="" search="" borowska+211a,+50-556+wroclaw,+poland?entry="gmail&source=g">
> > >
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org <mailto:amber.ambermd.org>
> > > http://lists.ambermd.org/mailman/listinfo/amber <http: lists.ambermd.org="" mailman="" listinfo="" amber="">
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> </http:></mailto:amber.ambermd.org></https:></amber.ambermd.org></carlos.simmerling.gmail.com></amber.ambermd.org></carlos.simmerling.gmail.com></amber.ambermd.org>
> _______________________________________________
> 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 Wed Sep 11 2024 - 01:00:02 PDT
