QM/MM - Built-in vs. GROMACS interface

[Léon Luntadila Lufungula]

Hi all,

Quick question: I plan on doing some QM/MM calculations in the coming months and I notice that there are two options to do this with CP2K: the built-in QM/MM method or interfacing CP2K with GROMACS. So I was wondering if there is any clear advantage of one over the other? If possible I would like to just use the built-in version as I have zero experience with GROMACS, but I would consider it if it would prove to be more advantageous w.r.t. functionality or performance.

All the best,

Léon

[Léon Luntadila Lufungula]

Hi all,

I guess asking a question on a Friday evening wasn't the best timing on my side. So hopefully this second message has more succes. So, if someone could elaborate on the differences/advantages/disadvantages between the two methods, I would greatly appreciate their help!

All the best,

Léon

[Josip Lovrić]

CP2K is not so good in MM. So unless you have something super simple in MM part go for the interface.

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/f904e1ea-558d-4e19-b645-ce224642bdf5n%40googlegroups.com.

[Léon Luntadila Lufungula]

Hi Josip,

Thanks for the clarification! I'm planning on simulating water at the MM level, so I guess that would require the interface right?

All the best,

Léon

[Josip Lovrić]

It is up to you to make evaluation, but from my experince I would rather choose interface.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/d173b65f-13fd-4066-aaec-920dc22119e8n%40googlegroups.com.

[Thomas Kühne]

Dear Leon, 

in my opinion the main advantages of Gromacs are the availability of tools to setup the MM part incl. 

topology and FF, easier usage in particular when you have previous grimaces knowledge and to a  

lesser extent performance. The latter, however, is for a QM/MM typically of smaller concern, since the 

relevant electrostatic interactions between the MM and QM parts are in both cases computed by the 

GEEP method of CP2K. 

So if you don’t have previous exposure to gromcas and only want to simulate solvent water by MM, 

this is pretty standard (at least in conjunction with typical fixed and flexible 3-body water models) and 

multiple examples available, e.g. https://www.cp2k.org/howto:biochem_qmmm

Greetings, 

Thomas Kühne

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/d173b65f-13fd-4066-aaec-920dc22119e8n%40googlegroups.com.