From: Francesco Pietra (
Date: Wed May 06 2020 - 02:49:55 CDT

Direct access to CGenFF results in carbocation ( and carbon radical or
not being supported. All servers relying on CGenFF, such as CHARMMGUI,
therefore report failure with such molecules. The same with MATCH server,
which does not rely on CGenFF.

I wonder whether any trick has been found to circumvent parameterization of
a carbocation (lying in a proteic or nucleosidic binding pocket) for the MM
(NAMD) part of QM-MM, leaving the QM task to ORCA. As the MM to QM linkage
does not occur directly to the carbocation, rather to the aminoacids or
nucleosides surrounding it, probably it should be possible. However I am
short of imagination how.

Thanks for advice
francesco pietra