Amber GB models in C++ API

[Charles Brooks]

Hello,

I have been looking at the additional features for Amber-like GB models in the python interface/API. I cannot find that these are available in other languages, is that correct? If not correct, please let me know how I would access them, as I don't find them documented. I'm most interested in whether the effects of finite salt concentrations and more recent (than OBC2) GB models are available as they are in the python API?

Thanks,

Charlie

[Peter Eastman]

Hi Charlie,

All of these GB models are implemented using CustomGBForce. The code we use to build them is written in Python, but that code is very simple and should be easy to port to another language. Take a look at https://github.com/SimTk/openmm/blob/ma ... bforces.py, and of course the documentation on CustomGBForce at https://simtk.org/api_docs/openmm/api6_ ... Force.html.

Peter

[luke czapla]

Hello,

I have been looking at the additional features for Amber-like GB models in the python interface/API. I cannot find that these are available in other languages, is that correct? If not correct, please let me know how I would access them, as I don't find them documented. I'm most interested in whether the effects of finite salt concentrations and more recent (than OBC2) GB models are available as they are in the python API?

Thanks,

Charlie

Hi Charlie,

I've ported a few to C++. It's mainly just rewriting the CustomGBForce routines in Python into the C++ syntax. I made a OBC2Force object (igb = 5 in AMBER and in customgbforce.py) and a GBN2Force object (igb = 8 in AMBER and in customgbforce.py) that I've been using for this problem with aggregation of micelles from random monomers. Looks like the GBn2 works a lot better for this system and I'd gotten stuck with what looked like a reverse micelle in a system that should form a micelle with OBC2 (and Mbondi2 radii). Dave Case's researcher suggested this might be analogous to the overstabilization of salt bridging but it is just a hypothesis since there's no net charge on these PEG-alkane copolymers.

best regards,
Luke