Re: [Open Babel] [OpenBabel-Devel] Default force field option for --gen3d

[Geoffrey Hutchison]

> Depending on the scenario, I would consider either vacuo (1) or water (80) to 
> be good 
> choices, but 4 is a 'close enough approximation' of 1.

There are a variety of papers suggesting that the effective dielectric constant 
inside a peptide is ~4.0. Most organic solvents are ~2-2.3 epsilon. 

Several other people have suggested that Open Eye has a MMFF94 variant with no 
electrostatics, but repulsive VdW interactions. That seems a bit strange, but 
essentially means that they feel the vacuum electrostatics are too strong.

In 3.0, the dielectric constant is adjustable via Python or C++ but I'd like a 
default, especially for the —gen3d operation, which is used heavily. In my 
tests, it doesn't have much effect, but YMMV.

The -forcefield command-line "op" already has control over epsilon as well:
https://github.com/openbabel/openbabel/blob/master/src/ops/forcefield.cpp

-Geoff

___
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss

Re: [Open Babel] [OpenBabel-Devel] Default force field option for --gen3d

[Stefano Forli]

I agree with the new MMFF95s default, most of our common cases are modeled well 
with 
planar aromatic nitrogens.

About the dielectric, I don't have strong opinions.
Depending on the scenario, I would consider either vacuo (1) or water (80) to 
be good 
choices, but 4 is a 'close enough approximation' of 1.

S

On 9/22/19 5:11 PM, Geoffrey Hutchison wrote:
> While it's possible to change the force field with the -gen3d option, most 
> people don't.
> 
> I think for 3.0 it might be nice to:
> - Use MMFF94s (i.e., the one with planar aromatic nitrogens)
> - Use a default dielectric constant of ~4.0
> 
> In my testing #1 helps a bit - and #2 at least doesn't hurt when comparing to 
> experimental crystal structures.
> 
> Thoughts?
> 
> -Geoff
> 
> ___
> OpenBabel-Devel mailing list
> openbabel-de...@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/openbabel-devel
> 

-- 

  Stefano Forli, PhD

  Assistant Professor
  Center for Computational Structural Biology

  Dept. of Integrative Structural
  and Computational Biology, MB-112A
  The Scripps Research Institute
  10550  North Torrey Pines Road
  La Jolla,  CA 92037-1000,  USA.

 tel: +1 (858)784-2055
 fax: +1 (858)784-2860
 email: fo...@scripps.edu
 http://www.scripps.edu/~forli/

___
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss