The motions of the hydrogens look pretty normal in my repeat of your
production run. But, I used sander.MPI in Amber 11. I'd have to
compile Amber 10 before I could try it there.
Could you quantify "a lot" with regard to the C-H stretching? I chose
a C-H bond at random and got an average of 1.09 with a standard
deviation of 0.0004. If you're getting something different from that
in general, let me know, and I'll try amber 10.
The rotations are pretty normal. The greater dielectric in solution
(implicit or otherwise) means that hydrogen bonding isn't as strong.
So, the hydroxyls do tend to rotate around a lot in solution. This is
one of the reasons the GLYCAM force field uses ensemble-averaged
charges rather than charges derived from a single structure: it tries
to model charge averaged over the normal motions of the hydroxyls. In
fact, all the hydrogens move a lot (except stretching, if all goes as
planned). The fact that you are in implicit solvent does make it
easier for things to move: there are no longer all those massive
waters that have to be jostled around before the solute can move. So,
they move around faster, but the basic movements should be similar.
If you are truly concerned about hydrogen bonding, though, you might
want to consider carefully your choice of implicit solvent model, and
whether you want one at all.
On Mon, Feb 20, 2012 at 11:08 AM, Lachele Foley (Lists)
<lf.list.gmail.com> wrote:
> The old style impropers should be fine. I don't see obvious problems
> with the structure either.
>
> I'm running your simulation now in an attempt to reproduce the
> situation. Will get back to you.
>
> On Mon, Feb 20, 2012 at 9:54 AM, <tommaso.casalini.mail.polimi.it> wrote:
>> Hi all,
>> I tried both with tleap and sleap (AmberTools 1.4) and nothing changes.
>> I performed a quick simulation in implicit solvent and the hydrogen
>> issues still remain.
>> I observed the structure that leap gives to me, but I was not able to
>> find the incorrectly placed hydrogens.
>> Moreover, tleap tells me that it must apply some old-prep improper
>> torsions, while sleap does not.
>> Thank you for your help.
>>
>> Tommaso Casalini, Ph.D fellow
>> CFA lab, Politecnico di Milano
>>
>> Def. Quota David A Case <case.biomaps.rutgers.edu>:
>>
>>> On Sat, Feb 18, 2012, Karl N. Kirschner wrote:
>>>>
>>>> I have encountered a similar problem recently, I don't think it
>>>> involves the glycam06g force field directly, but something with how
>>>> xleap (didn't try tleap) interprets the force field when sourcing the
>>>> protein/dna force field. If you look at the output structure of leap,
>>>> I think you will see that a few of the hydrogen atoms are incorrectly
>>>> placed. I corrected this by only loading glycam06 parameters and
>>>> suppressing the the protein force field loading (i.e. xleap -s -f
>>>> leap.in). This then gave me the correct spatial structure of the
>>>> carbohydrate.
>>>
>>> Hi Karl:
>>>
>>> Thanks for the info. Can you post a specific example? We need to track this
>>> down. If LEaP is actually adding hydrogens in incorrect ways, it should be
>>> fixed. (I doubt that there is any difference between tleap and xleap, since
>>> most of their code is shared.)
>>>
>>> Are there other codes that are more reliable in adding hydrogens to
>>> carbohydrates? "Reduce" does a good job for proteins, but I
>>> doubt(?) that any
>>> third-party program would know about all the atom and residue names used by
>>> GLYCAM. Either you or someone else might check sleap, since it would be
>>> helpful to know if the latter worked better than tleap/xleap.
>>>
>>> ...thanks!...dave
>>>
>>> (cc-ing to amber-developers as well)
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>
>>
>>
>> ----------------------------------------------------------------
>> This message was sent using IMP, the Internet Messaging Program.
>>
>>
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
>
>
> --
> :-) Lachele
> Lachele Foley
> CCRC/UGA
> Athens, GA USA
--
:-) Lachele
Lachele Foley
CCRC/UGA
Athens, GA USA
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 20 2012 - 09:30:01 PST
