For many reasons, we want an atomistic model for membranes. This is the issue for generating a simple atomistic / empirical force-field (GROMACS) model, probably with the Charm-GUI. (An alternative approach, using back-mapping from the coarse-grained Martini method is elsewhere.)