Re: tools for Charmm

From: Kenno Vanommeslaeghe (kvanomme_at_rx.umaryland.edu)
Date: Thu Feb 20 2014 - 13:59:02 CST

Hi Gianluca,

I'm probably telling you a lot of things you know already, but no general
force field has parameters for every single organic molecule it will
encounter. At best, it will have coverage of a comprehensive set of basic
chemical groups, parametrized in a generic context. It is implicitly
assumed that these generic parameters will keep on performing well if the
same chemical group is placed in an actual molecule, adjacent to other
chemical groups. However, in reality, this assumption of [parameter]
transferability is often problematic, especially when it comes to dihedral
parameters and charges in additive force fields. So for a significant
percentage of molecules, the force field can (and sometimes even needs to)
be made more accurate by optimizing the parameters specifically for that
molecule. This is a non-trivial thing to do, not only because it's
somewhat complicated, but also because a methodology should be followed
that is rigorously consistent with the rest of the force field, else
strange things may happen. Knowing this, we released a standard
methodology for use with the CHARMM (General) Force Field. That's the
"refinement" I was talking about in my e-mail re: "namd-l: tools for
Charmm" earlier today; all the information is in there. Further questions
about the parameterization procedure (but not about how to use it with
NAMD) can be directed to CHARMM's "Parameter Set Discussion" forum:
http://www.charmm.org/ubbthreads/ubbthreads.php?ubb=showflat&Number=29297

Cheers,

        Kenno.

On 02/20/2014 02:15 PM, Gianluca Interlandi wrote:
> Hi Kenno,
>
> I used CHARMM-GUI's Glycan Reader a few weeks ago. Extremely valuable! I'm
> just curious: do you mind sharing some tips about the "optimization" part
> that you mention?
>
> Gianluca
>
> On Thu, 20 Feb 2014, Kenno Vanommeslaeghe wrote:
>
>> Yes, CHARMM-GUI does use our CGenFF program in the back-end when it
>> encounters non-biomolecules. So it does parameter assignment. Though
>> some people use different terminology, we define parameterization :=
>> (parameter assignment + optimization). CHARMM-GUI does not do optimization.
>>
>> As an aside, CHARMM-GUI also has a unique and powerful function to
>> generate carbohydrate structures: Glycan Reader (but that one doesn't
>> generate parameters in the process).
>>
>>
>>
>> On 02/20/2014 01:08 PM, Gianluca Interlandi wrote:
>>> Have a look at
>>>
>>> charmm-gui.org
>>>
>>> On Thu, 20 Feb 2014, Peter Reinke wrote:
>>>
>>>> Hi all,
>>>>
>>>> Any suggestions of tools to parameterize molecules with Charmm force
>>>> field?
>>>>
>>>> Thanks for your attention,
>>>> Peter
>>>>
>>>>
>>>
>>> -----------------------------------------------------
>>> Gianluca Interlandi, PhD gianluca_at_u.washington.edu
>>> +1 (206) 685 4435
>>> http://artemide.bioeng.washington.edu/
>>>
>>> Research Scientist at the Department of Bioengineering
>>> at the University of Washington, Seattle WA U.S.A.
>>> -----------------------------------------------------
>>>
>>
>>
>
> -----------------------------------------------------
> Gianluca Interlandi, PhD gianluca_at_u.washington.edu
> +1 (206) 685 4435
> http://artemide.bioeng.washington.edu/
>
> Research Scientist at the Department of Bioengineering
> at the University of Washington, Seattle WA U.S.A.
> -----------------------------------------------------

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:10 CST