[gmx-users] Molecule removal by g_membed

Justin Lemkul jalemkul at vt.edu
Thu Aug 15 23:44:26 CEST 2013

On 8/15/13 3:05 PM, Bin Liu wrote:
> Hi All,
> I am simulating a system of lipopeptides embedded into a lipid bilayer. I
> have a question about how g_membed works.
> My bilayer has three kinds of lipids. Let's call them lipid_1, lipid_2,
> lipid_3. Sodium ions are used to balance the charges. When I invoke
> g_membed, it will ask for which group to embed in the membrane. In my case,
> obviously it is the group Lipopeptide. Then g_membed asks for a group to
> embed Lipopeptide into. Should it be the joint group of lipid_1, lipid_2
> and lipid_3? Or something else?
> I used make_ndx to generate an index group called Membrane by merging group
> Lipid_1, Lipid_2 and Lipid_3. It seems it works with g_membed. It gives the
> following information:
> Will remove 2 Lipid_1 molecules
> Will remove 1 Lipid_2 molecules
> Will remove 3 Lipid_3 molecules
> Will remove 7 SOL molecules
> Will remove 0 NA molecules
> Will remove 0 Lipopeptide molecules
> Apparently the behaviour of g_membed is what I want. But what concerns me
> is I don't know whether the removal of 0 Sodium ions is just a coincidence
> or it's a deterministic action by g_membed. Of course I don't want the
> removal of ions as it would cause a charged system. So my question is under
> what conditions will g_membed remove ions?

Whenever you delete molecules from a system, there will have to be corresponding 
changes in the topology, some that may require you to re-calculate the number of 
ions you need.  Molecules are deleted simply based on physical overlap.  Full 
details are in the g_membed paper (dx.doi.org/10.1002/jcc.21507).



Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441


More information about the gromacs.org_gmx-users mailing list