[gmx-users] grompp and [ ?types ]

[Erik Lindahl]

David L. Bostick wrote:

>Hello all,
>
>I just built a topology for a protein.  The termini were capped with ACE at
>the N and NAC (N-methyl) at the C.  One of the GLU residues is supposed to
>be GLUH.  I used pdb2gmx with all the appropriate options.  However, when
>doing a test grompp of the protein, I keep getting these kinds of warnings:
>
>WARNING 1 [file "//alberto/programs/gmx-3.1/share/clc_force_top/clcA.itp",
>line 27217]:
>  No default Angle types, using zeroes
>WARNING 2 [file "//alberto/programs/gmx-3.1/share/clc_force_top/clcA.itp",
>line 27219]:
>  No default Angle types, using zeroes
>WARNING 3 [file "//alberto/programs/gmx-3.1/share/clc_force_top/clcA.itp",
>line 27220]:
>  No default Angle types, using zeroes
>
>I checked these lines in the .itp files I created using the .top made by
>pdb2gmx.  They all coincidentally correspond to parameters in the ACE, NAC,
>or GLUH amino acids.  Furthermore, they are standard angletypes.. like
>
>HC C3 HC
>
>I looked for these in ffgmx2bon.itp and they are there.  I also added the
>parameters I found in ffgmx2bon.itp for these angles and I still get the
>error.  Why doesn't grompp find these in the .itp files it searches.
>

Hi David,

I suspect it has something to do with the atom naming...
Could you post the atom lines of those residues from the PDB file?  (One 
GLUH will do)

Cheers,

Erik