[gmx-users] multi-component membrane simulation

[Stanislaw Bobritsky]

Hello everyone!
It is given a lipid bilayer containing POPC (280 molecules), POPS (80) and
CHL(cholesterol, 40 molecules).
When I try to minimize this system in vacuo (steepest descent), I can see
the following:

Warning: 1-4 interaction between 19 and 34 at distance 4.892 which is larger
> than the 1-4 table size 2.400 nm
>
After that, of course, the system appears to be exploded.
(all topologies were generated with ProDrug server)

I tried to work with different component combinations. For example:

280 POPC:            Warning about 1-4 interaction, but system appears to be
alive. The 2nd minimization gives normal result with no warnings.
80 POPS:              The same.
40 CHL:                OK
1POPC+1CHL:     OK
1POPC+1POPS:   OK
280POPC+80POPS:    FAIL (warnings like previous, system destruction)

OK = no warnings, quick convergence.

280POPC+40CHL and 80POPS+40CHL boxes can be successfully minimized in the
same way. But if many POPC and POPS can be found in one box - the resulting
system is destroyed.

How can it be fixed?

Thanks for help!
Stanislaw Bobritsky,
ChemBio Center, Taras Shevchenko National University, Kiev, Ukraine.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20091012/e8d6dc0e/attachment.html>