"Partial potential fitting" with force matching

[Cecília Álvares]

Hello!

Quick question: suppose I have a system in its all-atom representation and I want to use it as reference to derive CG force-fields via force-matching. For that I have a trajectory file containing the position of all atoms and the force experienced by each of them.

The all-atom system has several atom types while its corresponding coarse-grained counterpart shall have only two superatom types (let's call them A and B) which are not bonded.

Let's suppose that for some reason I am only interested in the pairwise CG interactions A-A and A-B. Is it okay in the eyes of VOTCA if I declare in the fmatch.xml file only these two interactions? Would it give me correct potentials? What I mean is: will the algorithm, in the way it is implemented, give me the force-fields A-A and A-B that minimize the error of the overdetermined system of equations without considering at all the interaction B-B? (i.e., superatoms type B would exist in this context merely because I need to parametrize A-B and thus they will participate in that manner)

Or will the results VOTCA outputs be wrong for some reason?

Thanks in advance,
Cecilia

[Christoph Junghans]

It should work! Another way would be to subtract the B-B force from
the trajectory before doing forcematching. Similar to what we did
here: http://dx.doi.org/10.1002/mats.201100011

Christoph
>
> Thanks in advance,
> Cecilia
>
> --
> Join us on Slack: https://join.slack.com/t/votca/signup
> ---
> You received this message because you are subscribed to the Google Groups "votca" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to votca+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/7e119f2b-88b2-4686-815a-f0c6c1d6a565n%40googlegroups.com.



--
Christoph Junghans
Web: http://www.compphys.de

[Cecília Álvares]

Hmm.. now that I am ruminating about it, I dont think that it would work if I did things in the way that I said it.. because VOTCA will still include the superatoms type B in the system of equation: i.e., it will create equations Fi = blabla for the superatoms type B within the overdetermined system of equations... and, in these equations for superatoms type B, I would have only the force-field of A-B trying to live up to the total force Fi experienced by these superatoms (which is not true in practice since the B-B interactions still exist: I was just trying to not concern them in the fit and get the optimal A-B and A-A). 

Right? :///

I guess I would not only have to not declare hte B-B interaction in the fm.xml file but also I would need a way to delete Fi = blabla equations formed by superatoms i whose type is B, I guess.

I will check the method you mentioned carefully and I come back if the issue still exists..!

Thanks,

Cecilia

[Christoph Junghans]

On Wed, Aug 9, 2023 at 10:32 AM Cecília Álvares
<cecilia...@gmail.com> wrote:
>
> Hmm.. now that I am ruminating about it, I dont think that it would work if I did things in the way that I said it.. because VOTCA will still include the superatoms type B in the system of equation: i.e., it will create equations Fi = blabla for the superatoms type B within the overdetermined system of equations... and, in these equations for superatoms type B, I would have only the force-field of A-B trying to live up to the total force Fi experienced by these superatoms (which is not true in practice since the B-B interactions still exist: I was just trying to not concern them in the fit and get the optimal A-B and A-A).
> Right? :///
> I guess I would not only have to not declare hte B-B interaction in the fm.xml file but also I would need a way to delete Fi = blabla equations formed by superatoms i whose type is B, I guess.
>
> I will check the method you mentioned carefully and I come back if the issue still exists..!

Make sense, you can do a gromacs rerun with a topology where you
exclude all B-B forces and then run force matching on the new
trajectory like we did in that paper.

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/a08e6198-2263-4957-ac5f-3332e5f2f8ean%40googlegroups.com.

[Cecília Álvares]

Yeah, I will check! Thanks :)