Dear users,
I am trying to set up GROMACS 5.1 MD calculations for a protein-cofactor-ligand system.
After generating the cofactor and ligand topologies with acpype, I add them to the *.top file of the protein as you will find below.
I keep getting the same error message when I try to compile with grompp to add the ions:
Fatal error:
Syntax error - File PDB_ligand_GMX.itp, line 3
Last line read:
'[ atomtypes ]'
Invalid order for directive atomtypes
The funny thing is that when I follow EXACTLY the same procedure (using a bash script, so I really change nothing) with only the cofactor or only the ligand, it works. Can you tell me what am I doing wrong?
I guess I will have the same problem when it comes to add position restraints. Any advice before it happens?
Thanks everyone!
complex.top file
#################################################
; Include forcefield parameters
#include "myFF.itp"
; Include Ligand Topologies
#include "PDB_cof_GMX.itp"
#include "PDB_ligand_GMX.itp"
[ moleculetype ]
######### Unmodified text ##################
[ molecules ]
; Compound #mols
Protein_chain_B 1
PDB_cof 1
PDB_ligand 1
SOL 60
SOL 19505
http://www.gromacs.org/Documentation/Errors#Invalid_order_for_directive_xxx
Short answer: if your ligand introduces new [atomtypes], it must be #included before any [moleculetype] appears. That's why it works for a cofactor-in-water system; the cofactor is the first [moleculetype].
So, if I understand it well, grompp reads the directive [atomtypes] of cofactor.itp, then the [moleculetype] of the cofactor, then the rest of the cofactor.itp file and then it crashes when it tries to read the directive [atomtypes] of ligand.itp, is it?
So how can I do it if I have two *.itp files (one for the cofactor and one for the ligand)?
Thanks!
The entire force field must be established before any molecules can be defined. [atomtypes] is a force field-level directive. So if one .itp file introduces new [atomtypes] you need to put the #include statement immediately after the #include for the parent force field. If you have new [atomtypes] (or any bonded types) in multiple .itp files, you need to put them together in a separate .itp file and #include it after the parent force field (and then subsequently remove any of these parameters from the ligand or cofactor .itp files).
Thank you very much Justin. I add [SOLVED] to the title.
I would like to add that it is possible to have repeated atomtypes if you have, let's say, two ligands, so this must be checked and corrected before running grompp or it will crash with "WARNING: overriding atomtype XX"
Hi Joan,
I ran into the same error you encountered but I couldn't solve it. Reading through your discussion, I still don't understand how the syntax should work. Could you please post an example of the itp and top files that worked for you?
- Badges
- Science topic