[gmx-users] Re: Genion and number of Ions

[Marc Saric]

> It is quite easy to find the problem. Look at the .top file, in the
> rightmost column in the atoms section you will see qtot (total charge
> starting from atom 1). After each residue this number should be integer
> (although small round off errors may occur).
> 
> The problem is most likely a N- or C-terminus for which you selected

I let GROMACS do the autoselect.

I checked my chains (the *.itp-files), and did not find any error, all 
residues have integer charges.

Due to the fact that this is a bit tedious to do for ~ 1000 Residues, I 
wrote a small python-script (if anyone is interested...) to check (of 
course it finds wrong chargegroups if I insert some ;-) ).

The charge-error is reported by grompp, the itp-files look reasonable.

Any further idea? Should I simply ignore this and believe, that it is a 
rounding error? Does something might goe wrong further down the road 
(water-addition, compilation of EM-Input)?

I'm running GROMACS 3.1.4 on a SGI Origin with MIPS R12000 Processor, 
Chip Revision: 3.5, FPU: MIPS R12010 Floating Point Chip Revision: 0.0
and Irix 6.5. GROMACS has been compiled with single precision and fftw 
2.1.3 from fftw.org.

P.S.: Thanks for the help so far!

-- 
Bye,  Marc Saric

Max-Planck-Institut fuer molekulare Physiologie
Otto-Hahn-Str.11  44227 Dortmund  phone:0231/133-2168