Dear Xueqin Pang,
> 1. The input conformation is one molecular from the lipid membrane
> pdb (128 pre-stabled dppc pdb)
> (http://moose.bio.ucalgary.ca/index.php?page=Structures_and_Topologies
> dppc128.pdb), I just take coordinates of the first dppc molecular
> out and then add H atoms by Leap in amber. surprisingly, after
> optimizing by G03 (HF/6-31G* freq SCF=Tight Pop=MK IOp(6/33=2)),
With "HF/6-31G* freq SCF=Tight Pop=MK IOp(6/33=2)" I do not think you
do a geometry optimization, you get a frequency job for the Cartesian
coordinates provided (i.e. after a single point energy calculation).
To optimize a structure I would simply use:
HF/6-31G* Opt Test
or better
HF/6-31G* Opt=Tight Test
with "Freq"
(if you do want to check the frequencies _after_ geometry optimization)
HF/6-31G* Opt=Tight Freq Test
See
http://www.gaussian.com/g_tech/g_ur/k_freq.htm
> the tail of the DPPC changed a lot (please check the attachments).
> Actually, the conformation of different moleculars in the membrane
> pdb may be also different from others.
You "optimized" (did you really optimize your structure ?) in gas
phase a molecule that contains a positive and a negative charges. In
these conditions, this gas phase structure is different from your
initial/reference which I guess comes from condensed phase simulation.
Here, your problem is finding a way to select a conformation/a set of
conformations for a charged molecule taken out of its environment.
> 2.I get the RESP charges by RESP module of AMBER10 form Gauss output
> (antechamber -i g.out -fi gout -o test_resp.mol2 -fo mol2 -c resp)
> or by:
> (resgen -i test.ac -o test.respin1 -f resp1;
> respgen -i test.ac -o tesp.respin2 -f resp1;
> resp -O -i test.respin1 -o test.respout1 -e test.esp -t qout_stage1
> resp -O -i test.respin2 -o test.respout2 -e test.esp?-q qout_stage1
> -t qout_stage2
> antechanber -i test.ac -fi ac -o test_resp.ac -fo -c rc -cf qout_stage2)?
> So there is no selection of the conformation for MEP computation.
I do not use Antechamber that often; I cannot comment your commands &
I do not understand the "?" character after "test.esp?"
"no selection of the conformation": here, the selection of the
conformation is done by the Cartesian coordinates you provide in the
espot file (loaded with the -e option): I do not understand what you
mean.
> 3.I read the amber forum about calculating the atomic charges by
> RESP methods and now I am going to try R.E.D. But before moving to
> R.E.D., I still have some confusions to figure out.
If you do not find a way to select the conformation(s) you wish,
R.E.D. will not be more helpful than Antechamber. However, behind
R.E.D., R.E.D. Server and R.E.DD.B. all computational conditions used
in charge derivation have to be fully defined (conformation(s),
orientation(s), charge fit with minimum number of constraints, etc...).
> My questions are:
>
> a. Can I use the initial structure of DPPC to calculate the MEP and
> then fit the MEP to get atomic charges within R.E.D. G03 frequency
> calculation of the initial structure reveals three imaginary
> frequencies.
If you read
http://www.gaussian.com/g_tech/g_ur/k_freq.htm, it is said:
"Thus, it is meaningless to compute frequencies at any geometry other
than a stationary point for the method used for frequency
determination."
If you have "three imaginary frequencies" after geometry optimization
it is suspicious (in particular if your geometry optimization is
carried out without constraint).
> b. If the initial structure can't be adopted to evaluate the MEP, is
> the optimized structure reasonable for the task?
> In my opinion, the equilibrium structures of DPPC should not change
> very much during MD simulation, so the conformation used to derive
> the partial charges should be similar to the equilibrium structure.
> If the conformation changes too much after optimization, the atomic
> charges won't represent the electrostatic interactions during md
> simulaiton.
If I well understood your approach, your reference structure comes
from condensed phase simulation while your QM job is in gas phase; In
these conditions the final structure after geometry optimization (by
QM) should be different because your structure is doubly charged.
> c. If neither is correct, what parameters should be used to optimize
> the initial structure in order to calculate the MEP With restraints
> or taking the solvent effect into account?
I would summarize by:
- Is your structure really optimized by QM ?
- try to find a way to generate conformation(s) for your molecule that
makes sense.
Using the building block approach might help in your case.
See Cieplak et al. J. Comput. Chem. 1995, 16, 1357-1377.
If you want to use the building block approach, see examples .
http://q4md-forcefieldtools.org/Tutorial/
http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php
Amino-acid: with NH3+-terminal & (-)OOC-terminal fragment; charged fragments
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
Nucleic acid: anion fragments
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#25
Glycoconjugates: however only neutral form
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#26
These examples are available in R.E.DD.B. for download:
http://q4md-forcefieldtools.org/REDDB/projects/F-60/
http://q4md-forcefieldtools.org/REDDB/projects/F-82/
http://q4md-forcefieldtools.org/REDDB/projects/F-71/
Many new examples are going to be available...
There is also a Glycopeptide approach in R.E.DD.B. you could have used.
http://q4md-forcefieldtools.org/REDDB/projects/F-72/
Unfortunaltely, data are protected by the author.
We are going to release a statistics module in R.E.D.-IV allowing to
compare various set of charges (Mulliken, ESP, various RESP) when
using the building block approach in order to rigorously study the
impact of the constraints used during the fit. This should help to
improve the fit or at least to be aware of the errors introduced...
I hope this helps.
regards, Francois
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 06 2009 - 09:51:56 PDT
