Hello,
The next Amber release contains a fix for this. You need to add "keep_scr yes" into your TeraChem input file. For now, you should be able to do that by using the use_template option, which allows you to specify your own TeraChem input file.
Btw: the next Amber release will contain a new interface with TeraChem that enables faster QM/MM calculations. Stay tuned ;)
I hope this helps,
All the best,
Vinícius Wilian D. Cruzeiro, PhD
Postdoctoral Researcher
Department of Chemistry
Stanford University
Twitter: .vwcruzeiro
________________________________
From: 이욱(생명과학과) <lwmod.postech.ac.kr>
Sent: Friday, April 15, 2022 7:42 AM
To: amber.ambermd.org <amber.ambermd.org>
Subject: [AMBER] AMBER/TeraChem interface
Dear AMBER Users,
I am trying to run QM/MM MD with AMBER20(AmberTools21)/Terachem(Ver. 1.95Ab, trial version) interface,
The calculation goes well for the first MD step without any problem, but it gives an error in the second step as shown below (Terachem output file).
Can't open guess file scr/c0 for read
Please, fix "guess" keyword in your input.
DIE called at line number 3314 in file terachem/init.cpp
It seems that this is due to the mismatch between the actual name of a scratch directory created for Terachem and the name of the scratch directory given in Terachem input file.
As far as I understand, a scratch directory for Terachem is automatically created for each MD step, and its name is given like ./scr.inpfile.n-1, where n stands for step number. Therefore, the name goes like ./scr.inpfile for step 1, and ./scr.inpfile.1 for step 2…, but, in the input file for Terachem calculation created by AMBER/Terachem interface, the name of a scratch directory is always ./scr.inpfile. This is why the first MD step goes well, but an error occurs in the second MD step because it can’t find a wavefunction file (c0) located in ./scr.inpfile.1 directory in the second MD step.
I’ve tried to manually define the name of a scratch directory for Terachem, but it didn’t work.
Any help would be appreciated!
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 15 2022 - 09:00:02 PDT
