Re: MD simulation Patch grid errors

From: Jeffrey Potoff (
Date: Mon Jan 13 2014 - 22:34:42 CST

On 1/13/2014 10:37 PM, Jiann George Pazhamalil wrote:
> Hi,
> Thank you for your responses,
> The POPC membrane was made through charmmgui and then converted
> accordingly to work in the lipid11 amber force field. I am attempting
> to reduce the number of steps in the NAMD simulation and seeing if
> that would alter the patch grid problem. From the logout files it is
> visible that the system is shrinking at a high rate.
> I thought this could be due to the surfacetenstiontarget parameter as
> there are discrepancies between the units used in amber and namd. In
> the lipid11 Amber the units were dyn/Angstrom (the value I found and
> am using) and namd specifies it as dyn/cm but converting 17 to
> 17000000 doesn't seem right. From the other mailing lists everyone
> seems to use a value closer to 20. Is this parameter applying pressure
> from outside or from within the system?
No, that is not correct. If you check the figure caption of Figure 4 in
the Lipid 11 paper ( | J. Phys. Chem. B
2012, 116, 11124−11136), the units are listed as dyn/cm.

"Figure 4. Area per lipid versus simulation time for lipid bilayer. All
surface tension values are reported as dyn/cm per layer (abbreviated
“dyn” in the legends)."

That said, if one is intent on using AMBER, you may want to consider
using GAFFLipid, (Dickson et al. Soft Matter, 2012, 8, 9617), which has
updated parameters and does not require the use of a surface tension to
get reasonable area per lipid values.

Jeffrey J. Potoff               
Associate Professor and Director of Early Engineering Programs
Department of Chemical Engineering and Materials Science
Wayne State University			  5050 Anthony Wayne Dr
Detroit, MI 48202

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:02 CST