[gmx-developers] Question: Crystal symmetry support?

[Shirts, Michael (mrs5pt)]

> Problem is that you need at least one entire unit-cell, so you can not
> use symmetry to simulate only one protein instead of 8 in a P212121 cell.

Absolutely, agreed.  Note that with small molecules, this getting one (or
multiple unit cells wouldn't be that bad).

> As we support general triclinic cells, any crystal can be simulated.
> Simulating only one copy will almost never work, as this would
> introduce singularities.

OK, then perhaps we should be thinking about ways to automate the setup for
different space groups?  I'm happy to look at this (with others, if there is
additional interest), though that will be a bit lower on my priority list.

Best,
~~~~~~~~~~~~
Michael Shirts
Assistant Professor
Department of Chemical Engineering
University of Virginia
michael.shirts at virginia.edu
(434)-243-1821