Re: Protein in a Lipid Membrane

From: Ale Gomez (agomez.fisica_at_epn.edu.ec)
Date: Fri Jun 11 2010 - 08:40:33 CDT

Hi Basak
I really appreciate your help, thank you.
Actually I made another things before I equilibrate my system. First I
minimize the entire system and then warm it up to 300K, both steps without
any fixed atoms, and now I tried to equilibrate it.
I had some problems when I warm it up, but I solve it with these parameters:

# Force-Field Parameters
exclude scaled1-4
#1-4scaling 1.0
cutoff 12.
switchdist 10.
pairlistdist 14
margin 1

# Integrator Parameters
timestep 0.1 ;# 1fs/step
rigidBonds none ;# needed for 1fs steps
fullElectFrequency 4
stepspercycle 20

So, now I do not know If you think that I have to equilibrate first only my
lipid membrane and then the entire system anyway.
Thank you in advance.

-------------------------------------------------------------------
Ale Gómez
Biophysics and Molecular Modeling Group
Physics Department
Escuela Politécnica Nacional, Quito - Ecuador
Ladrón de Guevara E11-253.
Casilla 17-01-1253
http://www.ciencias.epn.edu.ec/~biomod/

On 10 June 2010 18:18, Basak Isin <isinbasak_at_yahoo.com> wrote:

> Hello Ale,
> These can help solving your problem:
> 1. Fixing everything in the system including the lipid head groups and then
> minimizing and equilibrating the tail of the lipids as described in the
> membrane tutorial. This will relax the tails around your protein.
> 2. Starting the simulations at a low temperature and gradually increasing
> it.
> 3. Increasing the margin (you can make it 3).
> 4. If nothing else works, you can use membranes that are already well
> equilibrated. There are some deposited at Peter Tieleman's website (
> http://moose.bio.ucalgary.ca/index.php?page=Structures_and_Topologies) but
> these are very short equilibrations and these lipids are for Gromacs and
> converting them to the NAMD format might be cumbersome. There might be other
> places to download equilibrated lipids.
>
> 0.1 fs is too short for step size.
>
> Good luck,
>
> Basak
> Note: I tried sending another email previously but my browser froze so you
> might have gotten a similar email before :o).
>
>
> Basak Isin
>
> ABOVE ALL NATIONS IS HUMANITY.
> "What matters - all that matters, really - is the will to happiness, a kind
> of enormous, ever-present consciousness. The rest - women, art, success - is
> nothing but excuses." - Albert Camus
>
>
> Cause life is meant for living. Yaki-da ;o)
>
>
> ------------------------------
> *From:* Ale Gomez <agomez.fisica_at_epn.edu.ec>
> *To:* namd-l_at_ks.uiuc.edu
> *Sent:* Thu, June 10, 2010 6:22:33 AM
> *Subject:* namd-l: Protein in a Lipid Membrane
>
> Hi everyone
> I am working on a system with a protein in a lipid membrane. My system has
> almost 40 thousand atoms and measures 75X75. I had a lot of problems
> preparing the equilibration conf file because it doesn't run (my system
> became unstable or has atoms moving too fast) unless I use the following
> parameters:
>
> # Force-Field Parameters
> exclude scaled1-4
> 1-4scaling 1.0
> cutoff 22.
> switching on
> switchdist 18.
> pairlistdist 24.5
> margin 1.5
>
> # Integrator Parameters
> timestep 0.1 ;# 2fs/step
> rigidBonds none ;# needed for 2fs steps
> nonbondedFreq 1
> fullElectFrequency 2
> stepspercycle 20
>
>
> Comparing with the parameters in membrane tutorial (available in NAMD
> website), my numbers are bigger and I do not know if these are correct.
> Could any one help me with this predicament??.
> Thanks in advance!!.
> -------------------------------------------------------------------
> Ale Gómez
> Biophysics and Molecular Modeling Group
> Physics Department
> Escuela Politécnica Nacional, Quito - Ecuador
> Ladrón de Guevara E11-253.
> Casilla 17-01-1253
> http://www.ciencias.epn.edu.ec/~biomod/>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:13 CST