> 1. Indeed, it was Dave Cerutti that presented the dihedral fitting.
> Chad, Dave and I discussed this quite a bit.
> 2. We did publish a paper with Chad on this, please see
> doi:10.1021/ci500112w
>
Can you send me the software for this please? We are trying to fit some really tricky torsion terms in some custom lipids and running into all sorts of issues with the standard genetic + simplex least squares fit. It would be really helpful to try your approach instead and see how it does here.
> I will expand a bit later on on my own thoughts on this discussion, but
> let me state clearly that it is PURELY semantic. Of course, using either
> functional form gives the exact same answer.
>
Yes it does - for MD - but for actual fitting of Vn it makes a difference (if you are fitting to energies). The correct approach, I believe is to fit to forces or better yet to the Hessian matrix - since the energy, for a multi-dimensional system (where the dimensions are the parameter terms) does not give a unique solution - I think.
All the best
Ross
/\
\/
|\oss Walker
---------------------------------------------------------
| Associate Research Professor |
| San Diego Supercomputer Center |
| Adjunct Associate Professor |
| Dept. of Chemistry and Biochemistry |
| University of California San Diego |
| NVIDIA Fellow |
|
http://www.rosswalker.co.uk |
http://www.wmd-lab.org |
| Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
---------------------------------------------------------
Note: Electronic Mail is not secure, has no guarantee of delivery, may not be read every day, and should not be used for urgent or sensitive issues.
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Tue Oct 20 2015 - 08:00:04 PDT
