ONE quantum atom moves away during prepare for a qmmm job


Just Got Here
Dear all,

My qmmm system has two groups inside a protein defined as quantum, a CO3 ion and a FE ion.

During the prepare job, the FE atom always moves away to weird coordinates, far away from the starting place, but the CO3 stays in place. Needless to say, in reality they stick together inside the protein.

Any ideas why this happens?

If I run a qmmm dft optimization on this crazy system, it converges ok, and the FE ion remains in the weird coordinantes.

I'm enclosing links to the script and source files, coulnd't figure out how to upload them here.

source_pdb
script_file

Thanks for the attention,
Pd

PS - I checked the pdb file format. it looks ok, even when I compare it with the N3o3.pdb file on nwchem documentation:
N3o3.pdb
ATOM      1  O1  NN3     1       1.562   0.180  -1.368    
ATOM      2  N1  NN3     1       1.316   0.936  -0.424    
ATOM      3  N2  NN3     1       2.364   1.026   0.488    
ATOM      4  O2  NN3     1       3.472   0.429   0.421    
ATOM      5  N3  NN3     1       1.958   1.906   1.488    
ATOM      6  O3  NN3     1       2.814   2.068   2.361    
source pdb file
ATOM      1  N   ASP A   3      33.430  27.007  61.687  0.0782 1.8240
ATOM      2  CA  ASP A   3      32.228  27.083  60.871  0.0292 1.9080
ATOM      3  C   ASP A   3      31.484  28.448  61.015  0.5621 1.9080
[...]
ATOM   5021  HA  CYS A 331      55.791  72.564  61.090  0.0938 1.3870
ATOM   5022  HB2 CYS A 331      55.200  75.268  60.409  0.1228 1.3870
ATOM   5023  FE  FE  A 400      49.913  50.743  51.393  1.0000 0.0000 
ATOM   5024  C   CO3 A 500      49.752  52.852  50.082  1.0000 1.0000
ATOM   5025  O1  CO3 A 500      50.624  52.623  50.963  1.0000 1.0000
ATOM   5026  O2  CO3 A 500      48.843  51.940  49.831  1.0000 1.0000
ATOM   5027  O3  CO3 A 500      49.755  54.000  49.449  1.0000 1.0000

Clicked A Few Times
Dear Pedro Deira,

Have you find where the problem came from ? If so could you help me please because I have got exactly the same problem. Weirdly it happens when I run QM/MM but it does not when I run MM only...

Many thanks,

Bastien


Forum >> NWChem's corner >> QMMM