The Molecular Graphics Laboratory Forum

Hello:
Good morning. In the recent past I encountered issues of unrecognized atoms by ADT 1.4.4. on GNU/LINUX with pdb files from Amber.

Now I came to same ADT with pdb files for a homotrimeric protein equilibrated with NAMD ff CHARMM. As for a "constant pH" simulation, negatively monocharged couples GLU-ASP and GLU-CL, as well as many disulfide bonds, exist. The above problem is exacerbated:

(1) Immediately warning that these atoms have no bond:

P:LYS95:HZ3 [actually regularly existing for these atoms index 1532 8295 15018]

P:ILE220:HG12 [actually regularly existing for these atoms index 3526 10289 17011]

accepting as "go on", new warning:

long series of non-integral charges, such as
LEU1 0.1629
PHE49: 0.0084
....
The warning ends with "correct the charges in the output file" How can that be accomplished?

The evaluated total charge -55.1037 is absurdly highly negative for this protein (as it was prepared). Apparently, ADT does not recognize protonations carried out and amenable to regular equilibration into a lipidic membrane (with excellent rmsd) with the CHARMM ff.

As I have ruled out any mistake in my files, this must be a problem from the babele of the formalisms.I would be happy to hear from the audience how they solve these problems, or have no such problems. Ideally, I would like to go to VINA or AUTODOCK docking simulations with my polar hydrogens (and mol2 charges).

Thanks
francesco pietra