Re: [AMBER] sander seeks for the .skf integral file for a single-empty atomtype

From: David A Case via AMBER <>
Date: Sat, 18 Feb 2023 08:39:01 -0500

On Fri, Feb 17, 2023, Maciej Spiegel via AMBER wrote:
>| 6 1 (S ) 6 () /net/software/testing/software/Amber/22.1-intel-2021b-AmberTools-22.3-updated-cpptraj/dat/slko/mio-1-1/S-.skf

You are correct: the problem is that the code thinks it has an atom with
an unassigned element.

> qmmask='.2135-2140 | :275 | .702-705 | .670-697',

Double/triple check your atom mask. The list of atoms that match the mask
should be in the MDOUT file, or you can use the ambmask utility. Are there
any extra points? It also could be that one of the atom types in your mol2
file never had a corresponding addAtom command in the frcmod file. I think
this might be allowed for MM calculations, but not for QM/MM.

> qm_theory='SCC-DFTB',

Another thing to try for debugging: set the method to something like pm6,
and run a one-step test calculation. You might get a different error
message, that could be more helpful.

....good luck...dac

AMBER mailing list
Received on Sat Feb 18 2023 - 06:00:02 PST
Custom Search