[gmx-users] Density of the box

[Mark Abraham]

On 10/01/2011 7:59 PM, kavya at rishi.serc.iisc.ernet.in wrote:
> Dear Gromacs users,
>
> I was trying to run a simulation using a protein of 123  aa long
> and wanted to adjust the density to 993.04g/l so with the following
> command I ran the editconf:
>
> editconf -f conf.pdb -o box.pdb -bt cubic -d 0.8 -density 997.0479
>
> Its output was as follows:
>
> --------------------------------------------------------------
> .
> .
> Read 1842 atoms
> Volume: 115.734 nm^3, corresponds to roughly 52000 electrons
> No velocities found
>      system size :  3.370  4.749  4.270 (nm)
>      diameter    :  4.979               (nm)
>      center      :  1.646  1.395  2.422 (nm)
>      box vectors :  4.707  6.445  3.815 (nm)
>      box angles  :  90.00  90.00  90.00 (degrees)
>      box volume  : 115.73               (nm^3)
> Volume  of input 115.734 (nm^3)
> Mass    of input 42476.2 (a.m.u.)
> Density of input 609.444 (g/l)
> Scaling all box vectors by 0.848671
> new system size :  2.860  4.030  3.624
>      shift       :  1.893  2.106  1.234 (nm)
> new center      :  3.289  3.289  3.289 (nm)
> new box vectors :  6.579  6.579  6.579 (nm)
> new box angles  :  90.00  90.00  90.00 (degrees)
> new box volume  : 284.70               (nm^3)
> ....
> ---------------------------------------------------------------
> My doubt here is it says that the Mass of input is 42476.2 (a.m.u.),
> but the mass of protein is 13795.4 Da (according to the protparam with
> sequence input). I am not sure whether protparam takes into account
> hydrogens. Even with the inclusion of hydrogens (885), the molecular
> weight will still be 14686.49Da. So how is 42476.2 a.m.u. comes?

Hmm, that's the second post in a week reporting editconf giving an 
apparently spurious measurement of the mass. Sounds like it could be a 
bug. Thoughts, anyone?

Mark