Re: [AMBER] Lipid14 POPE lipids

From: Gould, Ian R <i.gould.imperial.ac.uk>
Date: Fri, 20 Feb 2015 14:31:26 +0000

Hi Joe,

My first thought would be to run with the Berendsen barostat as that was
what we used for the paper and would then be consistent for comparison.
The MC barostat is newish and I¹d check to see what the acceptance ratio
of the moves you are getting with the default number of steps.
Ideally you should be seeing something around the 50% acceptance, if not
you need to adjust mcbarint to get closer to that value.
Also take a check on the initial structure that the charmm-gui generated
sometimes we¹ve seen some odd structures which can take a long time to
equilibrate out.
You may want to take a look at our recent paper DOI: 10.1039/c4cc09584g
where we perform self assembly of lipid bilayers with Lipid 14 and C36,
where we start from totally random lipid/water structures and observe for
3 microsecond long simulations good agreement with the area per lipid.
Again we used the berendsen barostat for this work.

HTH
Ian

"Wave after wave, each mightier than the last
'Til last, a ninth one, gathering half the deep
And full of voices, slowly rose and plunged
Roaring,and all the wave was in a flame"
Tennyson

-- 
Dr Ian R Gould, FRSC.
Reader in Computational Chemical Biology
Department of Chemistry
Imperial College London
Exhibition Road
London
SW7 2AY
E-mail i.gould.imperial.ac.uk
http://www3.imperial.ac.uk/people/i.gould
Tel +44 (0)207 594 5809
				
			
		
	
On 20/02/2015 12:42, "Joseph Baker" <bakerj.tcnj.edu> wrote:
>Hi all,
>
>I've been having one of my students run some test simulations of a pure
>POPE bilayer in pmemd.cuda with Amber14 using the Lipid14 force field. My
>student has been following the protocol in the Lipid14 tutorial online.
>Attached is a figure to show the current area per lipid vs time (in
>picoseconds) we are getting (plotted are the steps from the beginning of
>the series of Hold steps during which the box size equilibrates, and the
>production phase).
>
>I can explain why the green and red curves behave badly. For the green
>curve my student accidentally only restrained half of the lipids during
>the
>heating phase, we were also underhydrated (~ 26 waters/lipid instead of
>the
>suggested 32/lipid in the Lipid14 paper for POPE), and we were running at
>303 K (instead of 310 K as suggested for POPE in the Lipid14 paper). The
>red curve represents a simulation where we fixed the restraints step
>during
>heating, but were still underhydrated and at a lower T.
>
>The black curve has appropriate hydration (enough waters in the system to
>have at least 32 waters/lipid head), and we are now running at 310 K. It
>seems that we are still getting a couple of angstrom^2 smaller area than
>is
>reported in the Lipid14 paper however. I should note that our initial
>lipid
>system was built using charmm-gui, it consists of 256 lipids instead of
>the
>128 in the Lipid14 paper, and we are using the monte carlo barostat
>instead
>of the Berendsen barostat for production (so the same as in the tutorial,
>but different from what was done in the paper). My guess is that none of
>these factors are resulting in us being a little low on the area per lipid
>however.
>
>Any suggestions would be appreciated! Thanks.
>
>Joe
>
>--
>Joseph Baker, PhD
>Assistant Professor
>Department of Chemistry
>C101 Science Complex
>The College of New Jersey
>Ewing, NJ 08628
>Phone: (609) 771-3173
>Web: http://bakerj.pages.tcnj.edu/
><https://sites.google.com/site/bakercompchemlab/>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

Received on Fri Feb 20 2015 - 07:00:03 PST