On Mon, Feb 01, 2010, Jason Swails wrote:
>
> There is an option in antechamber to specify Amber atom types, though
> having never used it on nucleotides before, I don't know exactly how
> well it works (i.e. whether it will be compatible with the way Amber
> names nucleic acid atom types).
When one is making a modification to an existing residue (such as in your
case), it is generally recommended (by me!) to ask antechamber to use Amber
atom types, not gaff ones. In this way, all the unchanged parts (including
the connections between your new residue and its surroundings) will be the
same as in the standard nucleotides, and you just have to check the parameters
involving the O-methyl part.
It is also trickier to generate Amber files for a fragment (like a nucleotide
that is to be inserted in a DNA chain) than for a complete molecule (which has
no dangling bonds). Here's what I would do:
1. Make a "complete molecule" to use with antechamber. I would suggest
N9-methyl guanine with the O6 modification. Run this through antechamber
(use amber atom types) to get charges and force field parameters. (Or, use
R.E.D. to get charges.)
2. Using xleap, copy the existing guanine nucleotide to a unit with whatever
new name you want, then edit this new unit, manually creating the
needed modifications for O-methylation. Select all atoms, and go to "edit
selected atoms", where you can manually enter the charges you got from step
1. You may want to make small modifications to keep most of the atoms with
the same charges, and you will need to also make adjustments to make sure the
total charge is correct. Save the results as an off file.
3. Now you can load this off file along with the usual force fields, and load
whatever frcmod file you got from step 1.
...hope this helps....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 01 2010 - 08:30:02 PST
