AMBER Archive (2005)

Subject: Re: AMBER: Question about GAFF atom types

From: Herbert Georg (hcgeorg_at_if.usp.br)
Date: Mon Jun 20 2005 - 11:47:38 CDT


What are the CB and CM atom types? I haven't found them in the gaff.dat
parm file.

Ilyas Yildirim wrote:

>It is better to compare the structures and use similar atom types, then.
>First that came to my mind is the following atom types:
>
> H1
> | H? O
>H1-CT-H1 \ /
> \ CM-C
> \ / \
> N*-CB CM-H?
> / \ /
> / C-CM
>H1- CT-H1 | \
> | O H?
> H1
>
>The H? can be HA. I just checked out the atoms and which atoms they are
>bonded. For instance, O bonded to C is always "O" type, and the C
>bonded to this O is always "C" type. But as I said this is just a guess.
>The bong length, bond angle, potential barriers... and dihedral angles.
>These needs to be defined. If there are not any force field parameters for
>this structure defined in the literature, and if you want to make a good
>guess, I would suggest you to compare the structure with the known ones,
>and define the missing parameters. Good luck...
>
>On Fri, 17 Jun 2005, Herbert Georg wrote:
>
>
>
>>I've tryied antechamber, but it sets wrong atom types ("ca" as if it
>>were a normal aromatic ring) to the atoms in the quinone ring. I
>>suppose, by reading the article, that the correct atom types for the
>>inner carbon atoms in a quinone ring is "cc" and "cd". Thats why I'm
>>trying to do it by myself.
>>
>>Ilyas Yildirim wrote:
>>
>>
>>
>>>Dear Herbert,
>>>
>>>Check out the AMBER tut. #5 of the 2005 AMBER Workshop.
>>>http://www.psc.edu/biomed/training/workshops/2005/Amber/psc_workshop/Tutorial_five/index.htm
>>>
>>>This tutorial is about antechamber, which will automatically predict the
>>>bond types and the needed parameters. If you know any experimental results
>>>for your structure (such as the bond length, bond angle, etc.) you can
>>>modify the frcmod file which will be created by this tutorial. Other than
>>>this, it will be very hard to create your own parameters. The best way to
>>>do is to compare with the other structures, and use those parameters for
>>>your system. Hope this helps.
>>>
>>>Good luck...
>>>
>>>On Fri, 17 Jun 2005, Herbert Georg wrote:
>>>
>>>
>>>
>>>
>>>
>>>>In fact here N is more like an N+.
>>>>
>>>>Herbert Georg wrote:
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>Hi, I'm trying to find the best choice of parameters to apply to a
>>>>>sistem like
>>>>>
>>>>>
>>>>> CH3
>>>>> \
>>>>> N == quinone
>>>>> /
>>>>> CH3
>>>>>
>>>>>I'm specially worried about the N, which is an amine N and makes 2
>>>>>bonds with a quinone ring. I'm worried also about the quinone carbon
>>>>>with which the nitrogen bonds. I looked at the J. Comp. Chem. article
>>>>>and also inside the parameter's file, gaff.dat, but couldn't decide
>>>>>for the pair of atom types. For example If I set the N atom as "nh"
>>>>>and the C atom as "c", I noticed that there are no parameters for the
>>>>>bond c-nh. Any light on this would be wellcome.
>>>>>Thanks,
>>>>>Herbert
>>>>>-----------------------------------------------------------------------
>>>>>The AMBER Mail Reflector
>>>>>To post, send mail to amber_at_scripps.edu
>>>>>To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>-----------------------------------------------------------------------
>>>>The AMBER Mail Reflector
>>>>To post, send mail to amber_at_scripps.edu
>>>>To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>-----------------------------------------------------------------------
>>The AMBER Mail Reflector
>>To post, send mail to amber_at_scripps.edu
>>To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>>
>>
>>
>>
>
>
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu