Hi Max,
The places you saw atom types being checked for `C1`, `C2`, and `C3` are
from GB radii assignment algorithms taken from tleap. You'll also notice
that alongside these type name checks are also checks about the atomic mass
of that carbon being >=13, 14, and 15, respectively.
Indeed tleap handled UA (united atom) force fields this way -- C1 types
"swallowed" one hydrogen atom, C2 types swallowed 2, and C3 types swallowed
3. But such force fields have fallen out of favor, and I haven't heard of
a new one in Amber being worked on in at least the past 15 years.
I put in the extra checks in ParmEd for the mass because the assumption
that the C1/C2/C3 atom types would only be used for UA force fields can
only be guaranteed when operating completely within Amber (that is
converting from another program's file types may very well use those atom
type names -- and often do), so checking that the mass is at least equal to
the sum of the mass of the carbon and assumed hydrogen atoms allows me to
avoid false positives. And since UA force fields have no hydrogen atoms
attached to those carbons, h-mass repartitioning (which would lower the
masses and yield false negatives) doesn't apply to those atoms, anyway.
HTH,
Jason
On Thu, Apr 7, 2022 at 4:30 PM Maximilian Ebert <max.ebert.me.com> wrote:
> Hi everyone,
>
> In $AMBERHOME/AmberTools/src/parmed/parmed/tools/changeradii.py there is
> an if statement checking for the atom type C1, C2 or C3. I couldn’t find
> this atom types in the AMBER force fields. I found a definition in the
> united-atom carbon types (
> http://www.uoxray.uoregon.edu/local/manuals/biosym/discovery/General/Forcefields/AMBER.html
> <
> http://www.uoxray.uoregon.edu/local/manuals/biosym/discovery/General/Forcefields/AMBER.html>).
> Could someone tell me what carbon atoms we look for at this place?
>
> Thanks,
> Max
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 08 2022 - 07:00:02 PDT
