Hi Kiril,,

I was wondering if anything needs to be changed in "orca.sh" script before running?

I have placed "orca.sh" and sander input in the same directory and I am wondering which should I run first.

This is my input:
minimization of solvent
&cntrl
imin = 1, maxcyc = 10000,
ncyc = 200, ntx = 1,
ntwe = 0, ntwr = 500, ntpr = 50,
ntc = 2, ntf = 2, ntb = 1, ntp = 0,
cut = 8.0,
ioutfm=1, ntxo=2, ifqnt = 1,
/
&qmmm
qmmask = '@12407-12427',
qmcharge = 1,
spin = 1,
qm_theory = 'EXTERN',
qm_ewald = 0,
writepdb = 1,
/
&orc
num_threads = 24,
/

and in the Slurm script I specify: sander -O -i min.in -o min.out -p CUTC_solv.parm7 -c prod5.rst7 -r min.rst7

My sander and XTB are installed through conda and I call Orca as a preinstalled module.

At the moment sander runs default orca level and does not even take XTB in consideration. If I run bash orca.sh it makes directory with two files and errors out:

head: error reading ‘../’: Is a directory
tail: error reading ‘../’: Is a directory
grep: ../: Is a directory
head: cannot open ‘energy’ for reading: No such file or directory
head: error reading ‘../’: Is a directory
tail: cannot open ‘gradient’ for reading: No such file or directory
head: option requires an argument -- 'n'
Try 'head --help' for more information.
head: error reading ‘../’: Is a directory
awk: fatal: cannot open file `pcgrad' for reading (No such file or directory)

It would be great if you could suggest the best way to make this work.

Thanks,

Marko