[gmx-users] Re: abnormal flip-flop when embedding cholesterol into DOPC using CG

[H.J.Risselada]

Dear Ran,

Indeed, strictly speaking a smaller time step will give you a more 
precise intergration over phase-space. However, as the CG-model itself 
and theirfore its corresponding energy description are parametrized 
for only a certain range in time steps. The extra details which one 
gains
with a more precise intergration over the energy field by chosing a 
much smaller time step are beyond the actual "resolution" of the 
CG-model and therefore not physical.

.........About the flip-flop problem. Please be more precize. Describe 
what is a high rate in flip-flops and how long you have simulated. The 
change of DOPC in cholesterol will give a dramatic change in the area 
per lipid. What one sees is probably a system trying to equlibrate 
itself. Does the problem still continue, after lets say 50 ns ?

Good luck,

greetings

JElger







On Thu, 06 Dec 2007 10:17:42 +0100
  Ran Friedman <r.friedman at bioc.uzh.ch> wrote:
> Dear Xavier, GMX users,
> 
> Xavier Periole wrote:
>> One other thing. In the CG FF the time step is considered as part of 
>>the
>> force field. Meaning that you should not play with it too much.
>> 0.04-0.02 ps is the actual range where the force field is considered
>> to reproduce experimental values of various quantities used for
>> paraameterization of the FF. Outside this range you change the 
>>balance
>> between the different terms and thus would alter the quality of the 
>>FF.
> Can you elaborate on this a bit? Surely if one changes the 
>temperature,
> pressure, compressibility, threshold distances etc., one doesn't
> necessarily reproduces the experimental results. Too long timesteps 
>may
> also cause a problem. But why would it be wrong to use a shorter 
>time
> step here?
> 
> Ran.
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before 
>posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php