Re: AMBER: charge constraints in resp

From: Ilyas Yildirim <yildirim.pas.rochester.edu>
Date: Thu, 30 Mar 2006 19:05:35 -0500 (EST)

Dear Kenley,

You say you want to impose the condition that the first 20 atoms to have
a total charge of zero. I wonder what the total charge of this structure
is; is it neutral or not.

In the AMBER manual, for the Duan et al. 2003 force field, it is saying
that the phi and psi torsions are changed compared to the old force field.
So, I am guessing that the RESP protocol is not changed while calculating
the resp charges in Duan et al.

[Another thing; Duan et al. force field is doing a Multimolecular resp
charge fitting, not for just one structure. So, even though u might end up
with some charges, I am not sure if you can use this structure in a
protein. The reason why I am saying this is because there are joint points
when two residues are connected; meaning that the joint point has to have
the SAME charge for all of the residues. Otherwise, the force field
won't be complete. In your case, because u are not doing any
multimolecular resp fitting, u dont have a joint point, and therefore
you cannot use these charges in a protein which has this particular
residue/structure.]

Let me explain more on what I wrote in my prev. email. I checked out your
.pdb file. If your structure has a total zero charge, what u should do is
the following:
---------------- First Fit ---------------------
&cntrl

nmol = 1,
ihfree = 1,
ioutopt = 1,

&end
 1.0
Resp charges
 0 26
 7 0
 1 0
 6 0
 6 0
 8 0
 1 0
 6 0
 1 0
 1 0
 6 0
 7 0
 6 0
 1 0
 7 0
 6 0
 1 0
 1 0
 1 0
 6 0
 1 0
 6 0
 8 0
 1 0
 7 0
 1 0
 1 0
------------------------------------------------
-------------- Second Fit ----------------------
&cntrl

nmol = 1,
ihfree = 1,
ioutopt = 1,
iqopt = 2,
qwt = 0.001,

&end
 1.0
resp charges
 0 26
 7 -99
 1 -99
 6 -99
 6 -99
 8 -99
 1 -99
 6 0
 1 0
 1 8
 6 -99
 7 -99
 6 -99
 1 -99
 7 -99
 6 0
 1 0
 1 16
 1 16
 6 -99
 1 -99
 6 -99
 8 -99
 1 -99
 7 0
 1 0
 1 25
--------------------------------------------------
There are three groups in your structure, which need to be equivalenced.
These are -N-(H)2 [N24, H25, H26]; -C-(H)2 [C07, H08, H09]; and -C-(H)3
[C15,H16,H17,H18]. If you want to use that particular constrain (that the
first 20 atoms should have a total 'zero' charge), u should use that
constrain in the first fit, not on the second fit. The meaning of second
fit is to recalculate the charges of the equivalent -H atoms.

I have done multimolecular resp fitting on nucleic acids. In my project, I
have an extra 2 unnatural bases, and therefore, I had to follow the
Cornell et al., 95 resp fitting protocol, and do a multimolecular resp
fitting on 7 structure (guanine, cytosine, adenine, uracil, isoguanine,
isocytosine, and DMP). If Duan et al. 03 is using the resp charges, the
same resp protocol should be used in that force field, too.

Best,

On Thu, 30 Mar 2006, Kenley Barrett wrote:

> Dear Ilyas,
>
> Thank you very much for your reply. I've attached the pdb file of my
> structure. It's a methylated histidine with blocking groups added in order
> to simulate the environment of a protein.
>
> I agree that restraining atoms 1-20 to charge 0 when everything but the
> hydrogens is frozen seems nonsensical. I made this choice because I've been
> trying to duplicate the charge fitting procedure of Duan et al. 2003 (so
> that my library file will be compatible with the 2003 force field). In their
> article they state that in the second fitting "the charges of the terminal
> blocking groups and those of heavy atoms were fixed". Assuming that "heavy
> atoms" meant everything but hydrogen, I ran a fitting according to this
> description. But my initial attempt resulted in a charge of about .12 on the
> non-blocking-group atoms, which is problematic because these atoms need to
> have a charge of 0 in order to make a usable library file. So a colleague
> advised me to do the same procedure, except adding a constraint that the
> non-blocking-group atoms had to add up to 0.
>
> I am very confused at this point about how to accurately duplicate the Duan
> et al. method while obtaining a 0 charge on non-blocking-group atoms. I'm
> not sure I understand your advice that I should add the constraint in the
> first fit and not the second one--wouldn't that allow the charges to move
> away from a state in which they add up to 0? And regarding your advice that
> I should not freeze the carbon atoms, I am just concerned that that would
> not agree with the resp fitting used for the 2003 force field. Does anyone
> have any experience doing resp fitting when trying to create library files
> using the 2003 methods? Would it be okay to just use the resp fitting
> procedure automatically created by respgen (which differs from what is
> described in the Duan et al. article)?
>
> Please let me know any suggestions that you have. I'm very grateful for your
> thoughts.
>
> Kenley
>
>
>
> On 3/30/06, Ilyas Yildirim <yildirim.pas.rochester.edu> wrote:
> >
> > Hi Kenley,
> >
> > I think u should use the restraints on the first fit; not on the second
> > one. Most of the atom charges on the second fit are frozen; only the
> > H-atoms are recalculated. So, having constrainst on the first 20 atoms
> > to have a total zero charge does not have any meaning.
> >
> > But still, I am not sure if it is correct what u have for the second fit.
> > What is the structure? U equivalence the H-atoms which are bound to the
> > same carbon, but u froze the charge of the C-atom in the second fit. U
> > should recalculate the whole group in the second fit; both the C-atoms
> > and H-atoms that are needed to be equilivalenced. Can u send a .pdb
> > file of the structure u are dealing?
> >
> > Best,
> >
> > On Thu, 30 Mar 2006, Kenley Barrett wrote:
> >
> > > Dear AMBER community,
> > >
> > > I am having some difficulties with resp. I want to do a second-stage
> > resp
> > > fitting in which atoms 1-20 are restrained to have a net charge of 0.
> > > However, once I've put the resp charges in a prep file and loaded it
> > into
> > > xleap, the charge of these atoms comes out to be 0.1230, not 0.
> > >
> > > I created the constraints in my resp2 input file based on the peptoid
> > > example in /opt/local/amber8/examples/resp_charge_fit/. The &cntrl
> > namelist
> > > is the same as what is automatically created by respgen (I've discovered
> > > that ioutopt needs to equal 1 in order for AMBER to go through all the
> > steps
> > > to make a working prep file).
> > >
> > > Here are the first and second stage resp input files:
> > >
> > > First stage input file:
> > > First resp fitting for methylated histidine, following Duan et al.
> > > procedure.
> > >
> > > &cntrl
> > >
> > > nmol = 1,
> > > ihfree = 1,
> > > ioutopt = 1,
> > >
> > > &end
> > > 1.0
> > > Resp charges
> > > 0 26
> > > 7 0
> > > 1 0
> > > 6 0
> > > 6 0
> > > 8 0
> > > 1 0
> > > 6 0
> > > 1 0
> > > 1 0
> > > 6 0
> > > 7 0
> > > 6 0
> > > 1 0
> > > 7 0
> > > 6 0
> > > 1 0
> > > 1 0
> > > 1 0
> > > 6 0
> > > 1 0
> > > 6 0
> > > 8 0
> > > 1 0
> > > 7 0
> > > 1 0
> > > 1 0
> > >
> > >
> > > Second stage input file:
> > > Second fitting for methylated histidine, following Duan et al. procedure
> > > except
> > > restraining the non-blocking atoms to a total zero charge.
> > >
> > > &cntrl
> > >
> > > nmol = 1,
> > > ihfree = 1,
> > > ioutopt = 1,
> > > iqopt = 2,
> > > qwt = 0.001,
> > >
> > > &end
> > > 1.0
> > > resp charges
> > > 0 26
> > > 7 -99
> > > 1 0
> > > 6 -99
> > > 6 -99
> > > 8 -99
> > > 1 0
> > > 6 -99
> > > 1 0
> > > 1 8
> > > 6 -99
> > > 7 -99
> > > 6 -99
> > > 1 0
> > > 7 -99
> > > 6 -99
> > > 1 0
> > > 1 16
> > > 1 16
> > > 6 -99
> > > 1 0
> > > 6 -99
> > > 8 -99
> > > 1 -99
> > > 7 -99
> > > 1 -99
> > > 1 -99
> > > 20 0.0
> > > 1 1 1 2 1 3 1 4 1 5 1 6 1 7
> > > 1 8 1 9 1 10 1 11 1 12 1 13 1 14 1
> > > 15 1 16 1 17 1 18 1 19 1 20
> > >
> > >
> > > Here is the -t output of the second stage fitting:
> > >
> > > -0.030630 0.030630 0.002225 0.457386 -0.567234 0.115050 -0.136863
> > > 0.094573 0.094573 0.173194 -0.449827 0.022620 0.156302 0.161476 -
> > > 0.180110 0.099700 0.099700 0.099700 -0.302310 0.181925 0.335545 -
> > > 0.538195 0.053203 -0.593932 0.360673 0.260624
> > >
> > >
> > > Strangely, if you add up the charges of atoms 1-20, the net charge is
> > > actually 0.12208 rather than 0.1230--is this kind of difference normal
> > after
> > > reading information into XLeap? But regardless of whether the charge is
> > > 0.12208 or .1230, I can't use this--I need to find a way to make the
> > charges
> > > of atoms 1-20 equal 0.
> > >
> > > I would be very grateful for any suggestions. Thank you in advance for
> > your
> > > help.
> > >
> > > Sincerely,
> > > Kenley
> > >
> >
> > --
> > Ilyas Yildirim
> > ---------------------------------------------------------------
> > - Department of Chemisty - -
> > - University of Rochester - -
> > - Hutchison Hall, # B10 - -
> > - Rochester, NY 14627-0216 - Ph.:(585) 275 67 66 (Office) -
> > - http://www.pas.rochester.edu/~yildirim/ -
> > ---------------------------------------------------------------
> >
> > -----------------------------------------------------------------------
> > The AMBER Mail Reflector
> > To post, send mail to amber.scripps.edu
> > To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
> >
>

-- 
  Ilyas Yildirim
  ---------------------------------------------------------------
  - Department of Chemisty       -				-
  - University of Rochester      -				-
  - Hutchison Hall, # B10        -				-
  - Rochester, NY 14627-0216     - Ph.:(585) 275 67 66 (Office)	-
  - http://www.pas.rochester.edu/~yildirim/			-
  ---------------------------------------------------------------
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Apr 02 2006 - 06:10:14 PDT
Custom Search