[gmx-users] Lipid parameterization

Wed Jun 15 19:09:36 CEST 2016

On 6/15/16 12:27 PM, Christopher Neale wrote:
> Dear Justin,
>
> thank you for the clarification. I wonder, however, how "highly optimized"
> are all of the charmm36 lipid parameters for headgroups other than PC and PE?
> The JPCB 2010 paper doesn't mention PG, PS, or PA headgroups at all (as far
> as I could tell), and Klauda's Mol Sim. 2015 paper "Molecular dynamics
> simulations of palmitoyloleoylphosphatidylglycerol bilayers" cites the 2010
> paper as the source of the PG headgroup parameters. As far as I can tell, it
> was only in 2015 that we learned that the Charmm36 PG parameters are any good
> and the book is still out on PS and PA.
>
> It's not a pedantic question, since the difference between "core charmm36" PA
> and paramchem PA might not be as large as one would assume, depending on how
> the core charmm36 PA lipids were parameterized.
>

This is probably a better question to ask Jeff Klauda.  I don't know offhand. 
One could easily compare PA from ParamChem with PA from CHARMM36 and try it.  I 
phrased my answer specifically for two reasons:

(1) One should *not* try to merge part of a molecule that was created with 
CGenFF and part that is taken directly from C36.  The force fields are 
compatible, yes, but CGenFF has that all-important word "general" in its name, 
therefore parameters obtained are generalized based on a collection of molecules 
with shared features.  Therefore, a given dihedral will have compromise 
parameters that are derived from fitting based on multiple molecules.  For a 
system like a lipid, with highly coupled rotations, this *could* lead to really 
bad behavior.

(2) Again related to generality, when I say "highly optimized," I refer solely 
to the concept of deriving lipid parameters for lipids.  It's actually rather 
difficult to get alkyl properties right, so some generic alkyl dihedral taken 
from butane is not necessarily valid for dodecane.  So "highly optimized" does 
not imply that I'm saying "these parameters are right no matter what," rather 
that the parametrization protocol accounts for different hings.

-Justin

> Thank you, 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 Justin
> Lemkul <jalemkul at vt.edu> Sent: 14 June 2016 14:35:26 To:
> gmx-users at gromacs.org Subject: Re: [gmx-users] Lipid parameterization
>
> On 6/14/16 2:04 PM, Christopher Neale wrote:
>> You might see if CHARMM_GUI has parameters for it. This is not to say that
>> they will be amazing parameters, as I have not looked into their approach
>> to generate parameters and it may well be rational cut and paste, but they
>> do have parameters for loads of lipids. If they don't have parameters, you
>> might also consider asking them to put your HG on their to-do list. My
>> understanding is that they want to add useful moieties.
>>
>
> CHARMM-GUI has access to the full CHARMM force field, and the full complement
> of every lipid parametrized for it (which is indeed quite extensive).  If a
> particular molecule (lipid type) is missing, it will call ParamChem (the
> CGenFF program's web server) to generate a topology for the full molecule.
> As I said in my first reply, this would be suboptimal as most of the
> necessary parameters will already be in the highly optimized CHARMM36 lipid
> force field.  Without knowing what the OP wants to work with, it's hard to
> say further, but I expect most of the work will already be done, and one
> needs to address a suitable model compound or two to work out any new
> linkages.
>
> -Justin
>
>> 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 Mohsen
>> Ramezanpour <ramezanpour.mohsen at gmail.com> Sent: 14 June 2016 13:53:26 To:
>> Discussion list for GROMACS users Subject: [gmx-users] Lipid
>> parameterization
>>
>> Dear Gromacs users,
>>
>> I am trying to parameterize a lipid molecule with a weird headgroup :-) in
>> Charmm36 force field for doing simulation in GROMACS. Reading through
>> literature, I found that Swissparam, and Paramchem.org are two useful
>> tools to make the topology files automatically.
>>
>> However, both seems to be suitable for small drug-like molecules than
>> large biological molecules, like lipids. Swissparam is a mixture of charmm
>> parameters and MMFF and is not based on any specific CHARMM FF version.
>> Paramcharm has also recommended to not use it for biological molecules.
>>
>> Although lipids are biological molecules, but I was wondering what will
>> happen if we consider the "whole synthetic lipid" as a drug-like molecule.
>>
>>
>> In this case, we might be able to mix it with other lipids which have been
>> parameterized in Charmm36?
>>
>> Thanks in advance for your comments Cheers Mohsen -- *Rewards work better
>> than punishment ...* -- Gromacs Users mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send
>> a mail to gmx-users-request at gromacs.org.
>>
>
> -- ==================================================
>
> 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
>
> ================================================== -- Gromacs Users mailing
> list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a
> mail to gmx-users-request at gromacs.org.
>

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

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

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