Hi all,
I have some updates.
I noticed that problematic oxygens from ATP have ATOMIC_NUMBER -1 in
.parm7 file. I don't know why, but after manually correcting them to 8,
QM/MM seems to work properly. Does anybody know how could this happen? I
am also wondering if it could possibly influence classical MD?
Juraj
On 31. 01. 24 10:33, Juraj Dobias via AMBER wrote:
> Dear developers and users,
>
> I compiled ambertools23 with QUICK support and successfully run QM/MM
> calculation on my system. My system contains protein ATP molecule with
> two Mg ions and druglike ligand. ATP parameters were obtained from
> Richard Bryce site (amber.manchester.ac.uk) and ligand was
> parametrized by openff sage2.1. I did standard MD using amber22
> without any issues. QM/MM seems to be fine if I include ligand and Mg
> ions in QM region, but after including ATP in QM region calculation
> crashes with error:
>
> At line 184 of file
> /home2/jurajdobias/QUICK/amber22_src/AmberTools/src/quick/src/modules/quick_sad_guess_module.f90
> (unit = 212)
> Fortran runtime error: Cannot open file
> '/home2/jurajdobias/QUICK/amber22/AmberTools/src/quick/basis/DEF2-SVP.SAD/':
> Is a directory
>
> From .out file it seems, that 3 oxygens on last phosphate were not
> correctly identified as oxygens:
>
> QMMM: QM Region Cartesian Coordinates (*=link atom)
> QMMM: QM_NO. MM_NO. ATOM X Y Z
> QMMM: 1 5963 Mg 0.8115 5.9113 1.4435
> QMMM: 2 5964 Mg 2.7773 4.4252 -0.8689
> QMMM: 3 5965 ^@^@ 2.0451 4.5731 5.3015
> QMMM: 4 5966 P 1.1880 5.4871 4.5387
> QMMM: 5 5967 ^@^@ 1.5994 5.7518 3.1545
> QMMM: 6 5968 ^@^@ 0.7944 6.6874 5.2830
>
> My explanation is that amber is searching for basis set in
> DEF2-SVP.SAD/ directory, but for these oxygens, it is missing element
> information, therefore it is trying to open directory.
> I am attaching input and output files.
>
> Any help solving the issue would be greatly appreciated.
>
> Juraj
>
> _______________________________________________
> 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 Feb 02 2024 - 03:00:02 PST
