[gmx-users] g_density appears broken in version 5

[Christopher Neale]

Dear membrane simulators:

If you use g_density from gromacs versions 5 to 5.0.4, please be aware that the -center option may be wrong:
http://redmine.gromacs.org/issues/1698

(g_density is also wrong for NPT simulations in previous versions, but for other reasons)

This is in addition to the fact that I believe from looking at the v5 code that it does not, as it states, build the density function outward in either direction from the bilayer center (as it indicates that it does in the g_density -h output and as is suggested to be done in http://redmine.gromacs.org/issues/1168 ). Rather, the new code uses variable bin-widths and a constant array along the bilayer normal. So please double-check your output against your own other analysis scripts.

Thank you,
Chris.