[gmx-users] Decreased Area/Lipid for Lipid Bilayer from CHARMM-GUI

[Justin Lemkul]

On 6/13/16 7:27 PM, Nidhin Thomas wrote:
> Hello Dr. Neale,
>
> Thank you for the quick response.
>
> I running a new DPPC-POPA mixed bilayer system with 0.15 M KCl concentration now. But I have observed similar decrease in area per lipid half way through the equilibration. The are a per lipid decreases from 63 A^2 to 59.9A^2 during first NPT step.
>

Are you modeling the exact composition of a known bilayer that has reliable 
experimental values?

> I have also modeled a pure DPPC bilayer with 36 lipids in a leaflet. This model also has a similar problem of reduced area per lipid.
>

How much does it reduce, and how long is your simulation?

> I have been reading research papers and came across this paper:
>
> S. Jo, J.B. Lim, J.B. Klauda, and W. Im (2009)
> CHARMM-GUI Membrane Builder for Mixed Bilayers and Its Application to Yeast Membranes. Biophys. J. 97:50-58 <http://dx.doi.org/10.1016/j.bpj.2009.04.013>
>
> It mentions that we have to perform either NPAT or NP(g)T with a small surface tension for small lipid bilayer systems.
>

That information pre-dates the current version of the force field, which is 
CHARMM36.  Special tricks were needed under CHARMM27, but this is no longer the 
case with the modern force field for lipid bilayers.  Be sure to read the 
pertinent publications about the force field and CHARMM-GUI itself.

dx.doi.org/10.1021/jp101759q
dx.doi.org/10.1021/jz200167q
dx.doi.org/10.1021/acs.jctc.5b00935

-Justin

> As mentioned earlier, in my system, I never used surface tension values. If I provide a non-zero surface tension value, would it change the properties of the system compared to systems simulated using other forces fields?
>
> Thanks a lot for the help!
>
> Nidhin Thomas
>
>
>
> Message: 4
>> Date: Mon, 13 Jun 2016 17:51:43 +0000
>> From: Christopher Neale <chris.neale at alum.utoronto.ca>
>> To: "gromacs.org_gmx-users at maillist.sys.kth.se"
>> 	<gromacs.org_gmx-users at maillist.sys.kth.se>, "gmx-users at gromacs.org"
>> 	<gmx-users at gromacs.org>
>> Subject: Re: [gmx-users] Decreased Area Per Lipid for Lipid Bilayer
>> 	obtained	from CHARMM-GUI
>> Message-ID:
>> 	<SN2PR03MB23010A1EA13D95F4EC7877C1C5530 at SN2PR03MB2301.namprd03.prod.outlook.com>
>> 	
>> Content-Type: text/plain; charset="Windows-1252"
>>
>> In future posts, it is best to copy and paste your .mdp file because we can not all get attachments from the list.
>>
>> Regarding your results, it is my understanding that the CHARMM-GUI has parameters for, and can build, many lipids systems whose parameters have not been completely validated by simulations in the literature (e.g. for things like APL). Last time I looked, PG was the only anionic charmm36 lipid that was shown to behave properly in bilayers (Klauda), though others may have since been shown to work well.
>>
>> If the issue is not simply that the PA parameters are not very good, then one likely source of problem is that perhaps you are not using the NBfix ions? That could be a big deal. I never use Na with membranes for this reason, so you could try KCl instead.
>>
>> Chris.
>>
>> ________________________________________
>> From: gromacs.org_gmx-users-bounces at maillist.sys.kth.se <gromacs.org_gmx-users-bounces at maillist.sys.kth.se> on behalf of Nidhin Thomas <nidhin.thomas0624 at gmail.com>
>> Sent: 12 June 2016 18:38:43
>> To: gromacs.org_gmx-users at maillist.sys.kth.se
>> Subject: [gmx-users] Decreased Area Per Lipid for Lipid Bilayer obtained        from CHARMM-GUI
>>
>> Hello GROMACS Users,
>>
>> I have modelled a mixed bilayer system consisting of 35 DPPC lipids and one POPA lipid (in each monolayers) using CHARMM-GUI. NPT ensemble was chosen while creating the model. TIP3 water model was used for modeling. I did not use surface tension for the membrane.
>>
>> I have simulated the system for 50 nanoseconds now. I have noticed that average area per lipid of the system is only 58.55 Angstrom2. I saw from literature that it should be around 63 and even at CHARMM-GUI, it is mentioned as 63 while creating the bilayer.
>>
>> I have used the exact .mdp files given by CHARMM-GUI for minimization, equilibration and production run.
>>
>> Can anyone please help me identify what could have possibly gone wrong in my simulation? I have attached the .mdp file used for production run and equilibration here for reference.
>>
>> In addition, I keep on getting these NOTES while running the grompp command prior to mdrun before it creates a .tpr file. I don?t know if this has got something to do with the simulation.
>>
>> NOTE 1 [file step7_production_2.mdp]:
>>  leapfrog does not yet support Nose-Hoover chains, nhchainlength reset to 1
>>
>> Estimate for the relative computational load of the PME mesh part: 0.05
>> Excluding 2 bonded neighbours molecule type 'TIP3'
>>
>> There was 1 note
>> Excluding 1 bonded neighbours molecule type 'POT'
>> Removing all charge groups because cutoff-scheme=Verlet
>> Setting the LD random seed to 2417118902
>> Estimate for the relative computational load of the PME mesh part: 0.05
>>
>> There was 1 note
>> Removing all charge groups because cutoff-scheme=Verlet
>>
>> Thanks a lot for the help!
>>
>> Nidhin Thomas

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================