This method should work if you treat the entire protein like the complex and
the receptor, with no ligand. Something like this should go in your .DECOMP
section (with COMPLEX=1, RECEPTOR=0, and LIGAND=0 in the .GENERAL section):
DCTYPE 4 # DCTYPE = 3 could also work here
#
COMREC 1-493
COMLIG 0
COMPRI 60-78 304-310
RECRES 1-493
RECPRI 60-78 304-310
RECMAP 60-78 304-310
LIGRES 0
LIGPRI 0
LIGMAP 0
This defines the entire system to be 493 residues and will print the
pairwise interaction energies for residues 60-78 and 304-310. I didn't know
what residues the auto-inhibition loop was interacting with in domain C, so
I just picked residues 304-310 as an example of the residues in domain C
that you would want to know the interaction energies for. You can substitute
the correct numbers here. In my experience, though, this should give you the
numbers you want for a system that lacks a ligand.
Good luck!
-Bill
2010/5/19 YuZhihong <comfort_79.hotmail.com>
>
> Hi, Bill
>
>
>
> Thanks for your suggestions! Yes, there is only one true molecule with A,B
> and C domains in my simulations. I've tried as you said, but it looks like I
> must set the residue numbers for all COM*, REC* and LIG* in the .DECOMP
> section, even I set RECEPTOR and LIGAND to 0 in the .GENERAL, otherwise the
> mm_pbsa.pl will not work. My protein has 493 residues, the auto-inhibition
> loop is from 60 to 78, then do you think what numbers I should set for COM*,
> REC* and LIG* in the .DECOMP? I have tried many numbers, but all of them
> were failed.
>
>
>
> Zhihong
>
>
>
>
>
> > Date: Tue, 18 May 2010 20:05:59 -0400
> > Subject: Re: [AMBER] Can I get interaction energy between two motifs in
> diffrent domains in the same protein through MD simulations?
> > From: brmilleriii.gmail.com
> > To: amber.ambermd.org
> >
> > If I understand your question correctly, there is only one true molecule
> > with three different domains, thus the idea of 'ligand' and 'receptor' is
> > not really applicable in your case. However, you can do a calculation
> with
> > mm_pbsa.pl in which you only perform a calculation on a complex (setting
> > COMPLEX equal to 1, and RECEPTOR and LIGAND to 0 in the general section).
> > Then you can choose to do a pairwise per-residue decomposition
> calculation
> > (setting DC to 1 and DCTYPE equal to 3 or 4). This will give you the
> > interaction energy for each residue in the complex interacting with every
> > other residue in the complex. Furthermore, you can choose to only print a
> > subset of the residues in the complex (i.e. only the residues in the loop
> of
> > domain A and few residues in domain C you are concerned with) by
> specifying
> > those residues with the variable COMPRI in the .DECOMP section.
> >
> > I hope that answers your question(s). Good luck!
> >
> > -Bill
> >
> > 2010/5/18 YuZhihong <comfort_79.hotmail.com>
> >
> > >
> > > Dear AMBER users,
> > >
> > > Now I'm studying a protein by MD simulations, this protein have three
> > > domains of A, B and C, one loop in A (call auto-inhibition-loop, AIL)
> > > interacted with a few residues in C and formed a so-called
> “auto-inhibition”
> > > state. I introduce a mutation at one interacting site in C and have
> finished
> > > 20ns MD simulations for both wild-type and mutant. Now I want to get
> the
> > > interaction energy of AIL with those residues in domain C from both MD
> > > trajectories, here is some questions:
> > >
> > > 1). Can I achieve the objective based on already finished classical MD
> > > simulations through some post-process, such as MM_PBSA/GBSA?
> > >
> > > 2). If I can get this type of interaction energy through MM_PBSA/GBSA,
> then
> > > how will I set input values in mm_pbsa.in, especially for
> > > COMPT/RECPT/LIGPT in .GENERAL section, and corresponding residue
> numbers in
> > > .DECOMP section? Supposing the organization of this protein is like:
> > > Domain A: 1 ~ 100
> > > AIL: 60 ~ 70
> > > Domain B: 101 ~ 200
> > > Domain C: 201 ~ 600
> > >
> > > 3). To get the total interaction energy of AIL with nearby residues in
> > > domain C, which type of energy decomposition is better for summing up,
> > > per-residue or pairwise per-residue? Which value is the best for
> DCTYPE?
> > >
> > > 4). If I can’t get this interaction energy based on finished MD
> > > simulations, can I get it by running other MD simulations in Amber9?
> Then
> > > which key parameters should I modify?
> > >
> > > 5). Even if the answer of 4) is “No”, could you give me some
> suggestions on
> > > how to compare the “binding affinity” of AIL to domain C in wild-type
> and
> > > mutant through computational technique?
> > >
> > > Thanks in advance, and I really, greatly appreciate any advice!
> > >
> > >
> > >
> > >
> > >
> > > Zhihong
> > >
> > > _________________________________________________________________
> > > 约会说不清地方？来试试微软地图最新msn互动功能！
> > > http://ditu.live.com/?form=TL&swm=1
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> >
> >
> >
> > --
> > Bill Miller III
> > Quantum Theory Project,
> > University of Florida
> > Ph.D. Graduate Student
> > 352-392-6715
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
>
> _________________________________________________________________
> MSN十年回馈，每位用户可免费获得价值25元的卡巴斯基反病毒软件2010激活码，快来领取！
> http://kaba.msn.com.cn/?k=1
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Bill Miller III
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-6715
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed May 19 2010 - 21:00:04 PDT
