[gmx-users] umbrella sampling and PMF calculation by using pull-geometry=direction in gromacs 5.0.4

Christopher Neale chris.neale at alum.utoronto.ca
Mon May 4 15:28:57 CEST 2015


Can you walk me through that image and tell me what you see that is different from what you expect? I see a maximum at z=0 in both cases (I presume z=0 is the bilayer center). I'm happy to continue guessing, but you are really best to take something like hexane in a simplified bilayer and do a test. If you don't see something like this: http://pubs.acs.org/doi/abs/10.1021/ja0535099 then you've got a problem.

________________________________________
From: gromacs.org_gmx-users-bounces at maillist.sys.kth.se <gromacs.org_gmx-users-bounces at maillist.sys.kth.se> on behalf of mahender singh <pharmbiochem at live.com>
Sent: 04 May 2015 00:57
To: gromacs user_list
Subject: Re: [gmx-users] umbrella sampling and PMF calculation by using pull-geometry=direction in gromacs 5.0.4

Thanks Chris for your help

 I am using Isoniazid drug and the membrane model is a mixture of different lipid types (POPE,POPC,POPI,POPS,cholesterol) in a ratio mimicking the plasma membrane more closely as compared to the simple single lipid type membrane model.  I have attached the PMF plot (attachment, shared link), and sorry for the partial PMF graph as one can see sampling is not sufficient , as I stopped further umbrella sampling, because of strange results. One is made by using pull-geometry=distance, free, energy on the negative side (limited to the one leaflet, as distance between two group become zero drug will not move onto the other side) and other by using pull-geometry=direction for the same drug i.e. isoniazid, free energy on the positive side.
https://onedrive.live.com/redir?resid=DA38458F80BE68E9!192&authkey=!ACGecvNH6tPzIlo&ithint=folder%2c

All the parameters were same except pull-geomerty i.e distance or direction. But free energy graph is completely opposite in both the cases and this should not be , as I thinks. As for the same drug free energy profile should not be completely opposite for the one drug, so may be I making some mistake in the umbrella sampling parameters. Can you provide me some suggestion on this topic.

Thanks in advance for help.

with regards
Mahender Singh



> Hello
> good morning to all
>
> I did steered md simulation by using following code
> ;pull code
>  pull        = umbrella
>  pull_geometry    = direction
>  pull_dim    = N N Y
>  pull_coord1_vec    =0 0 1
>  pull_start    = yes
>  pull_ngroups    =2
>  pull-ncoords    =1
>  pull_coord1_groups    = 1 2
>  pull_group1_name    = NPROT_ref
>  pull_group2_name    = LIG
>  pull_coord1_rate    = 0.004        ; 0.004 nm per ps = 4 nm per ns
>  pull_coord1_k        = 1000        ; kJ mol^-1 nm^-2
>  pull_nstxout    = 500        ; every 1 ps
>  pull_nstfout    = 500        ; every 1 ps
>
> And then used these results of SMD for umbrella sampling (total umbrella window =30 across the membrane) by using following code
>
>  pull        = umbrella
>  pull_geometry    = direction
>  pull_dim    = N N Y
>  pull_coord1_vec    =0 0 1
>  pull_start    = yes
>  pull_ngroups    =2
>  pull-ncoords    =1
>  pull_coord1_groups    = 1 2
>  pull_group1_name    = NPROT_ref
>  pull_group2_name    = LIG
>  pull_coord1_rate    = 0.0
>  pull_coord1_k        = 1000        ; kJ mol^-1 nm^-2
>  pull_nstxout    = 500        ; every 1 ps
>  pull_nstfout    = 500        ; every 1 ps
>
> In my case, I am moving a molecule from -z to +z across the membrane. So I used pull-geometry=direction in gromacs 5.0.4 (as per gromacs site= http://www.gromacs.org/Documentation/How-tos/Potential_of_Mean_Force), as I cannot use pull-geimetry=distance. Every thing completed correctly (seems to me), but when I used gmx wham on the pullf.xvg files, PMF profile was on the positive side. PMF increased from 0 (bulk water) to 60 at the center of the membrane and than again decreased to the 0 (bulk water on the other side).
> I all the literature, PMF values for the molecule crossing a membrane is on the negative side and minimum at the center.
> I am confused about the result. Did I do something wrong with the parameters or understanding the PMF curve wrongly.
>
> can any one give their suggestion on this.
>
> thanks in advance of user time and help.
>
> with regards
> Mahender Singh
>
> --
> 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.
>
>
> ------------------------------
>
> Message: 3
> Date: Sun, 3 May 2015 12:25:01 +0000
> From: Christopher Neale <chris.neale at alum.utoronto.ca>
> To: "gmx-users at gromacs.org" <gmx-users at gromacs.org>
> Subject: Re: [gmx-users] umbrella sampling and PMF calculation by
>       using pull-geometry=direction in gromacs 5.0.4
> Message-ID: <1430655898617.32802 at alum.utoronto.ca>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Better yet, repeat something that is already out there in the literature. PMFs for water, lipids, and some amino acid side chains across the bilayer have been tested extensively.
>
> 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 mahender singh <pharmbiochem at live.com>
> Sent: 03 May 2015 01:07
> To: gromacs user_list
> Subject: [gmx-users] umbrella sampling and PMF calculation by using pull-geometry=direction in gromacs 5.0.4
>
> Hello
> good morning to all
>
> I did steered md simulation by using following code
> ;pull code
>  pull        = umbrella
>  pull_geometry    = direction
>  pull_dim    = N N Y
>  pull_coord1_vec    =0 0 1
>  pull_start    = yes
>  pull_ngroups    =2
>  pull-ncoords    =1
>  pull_coord1_groups    = 1 2
>  pull_group1_name    = NPROT_ref
>  pull_group2_name    = LIG
>  pull_coord1_rate    = 0.004        ; 0.004 nm per ps = 4 nm per ns
>  pull_coord1_k        = 1000        ; kJ mol^-1 nm^-2
>  pull_nstxout    = 500        ; every 1 ps
>  pull_nstfout    = 500        ; every 1 ps
>
> And then used these results of SMD for umbrella sampling (total umbrella window =30 across the membrane) by using following code
>
>  pull        = umbrella
>  pull_geometry    = direction
>  pull_dim    = N N Y
>  pull_coord1_vec    =0 0 1
>  pull_start    = yes
>  pull_ngroups    =2
>  pull-ncoords    =1
>  pull_coord1_groups    = 1 2
>  pull_group1_name    = NPROT_ref
>  pull_group2_name    = LIG
>  pull_coord1_rate    = 0.0
>  pull_coord1_k        = 1000        ; kJ mol^-1 nm^-2
>  pull_nstxout    = 500        ; every 1 ps
>  pull_nstfout    = 500        ; every 1 ps
>
> In my case, I am moving a molecule from -z to +z across the membrane. So I used pull-geometry=direction in gromacs 5.0.4 (as per gromacs site= http://www.gromacs.org/Documentation/How-tos/Potential_of_Mean_Force), as I cannot use pull-geimetry=distance. Every thing completed correctly (seems to me), but when I used gmx wham on the pullf.xvg files, PMF profile was on the positive side. PMF increased from 0 (bulk water) to 60 at the center of the membrane and than again decreased to the 0 (bulk water on the other side).
> I all the literature, PMF values for the molecule crossing a membrane is on the negative side and minimum at the center.
> I am confused about the result. Did I do something wrong with the parameters or understanding the PMF curve wrongly.
>
> can any one give their suggestion on this.
>
> thanks in advance of user time and help.
>
> with regards
> Mahender Singh
>
> --
> 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.
>
>
> ------------------------------
>
> Message: 4
> Date: Mon, 4 May 2015 01:07:16 +1200
> From: James Lord <jjamesgreen110 at gmail.com>
> To: "gmx-users at gromacs.org" <gmx-users at gromacs.org>,  Justin Lemkul
>       <jalemkul at vt.edu>
> Subject: Re: [gmx-users] Justin biphasic tutorial with controlled
>       adsoption       of protein
> Message-ID:
>       <CAAUhNntFBtGw92t_c98FqkhvWycDT453jVa_1Wr1fdM4mf1dDw at mail.gmail.com>
> Content-Type: text/plain; charset=UTF-8
>
> Hi Justin,
> Thanks for the info, your assumption re long axis of the box is correct
> (z), here is the .gro file I have for one the oil (decane)-protein (the
> protein is in the middle part of the free space above the oil surface). Is
> that what you were talking about? I went through manual and could not get
> what to do for position restrains, appreciate if you can elaborate further
> on it? Thanks for your final point re solvent/ions separate coupling.
> https://drive.google.com/file/d/0B0YMTXH1gmQsYmQzRXUtN1ZENmc/view?usp=sharing
> Cheers
> James
>
>
> On Sun, May 3, 2015 at 4:23 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
> >
> >
> > On 5/2/15 10:02 AM, James Lord wrote:
> >
> >> Dear gmx users,
> >> I have a biphasic system, like Justin's tutorial, but I want to control
> >> the protein to adsorb/desorb from the oil-water interface, I don't want
> >> the
> >> protein to go through oil phase just want to keep it at the interface, I
> >> have added a constant force at the end of the mdp file. i tried different
> >> pull_k1 values but what I see is, when the force is applied the oil is
> >> also
> >> deformed (which makes sense) and kinda make a hole in the oil phase and
> >> eventually pass through it. I just want to keep it at or away from the
> >> interface without deforming/puling away the oil molecules, What is the
> >> best
> >> way to do it? any suggestion? here is the mdp
> >> https://drive.google.com/open?id=0B0YMTXH1gmQsZ1ljOUZROWNGM3M&authuser=0
> >>
> >>
> > I wouldn't use the pull code.  This sounds like a task best suited for a
> > flat-bottom restraint.  I'm going to assume the long axis of the box is
> > along z for this example. You can create a coordinate file in which the
> > z-coordinate of all the protein atoms is equal to the middle of the box
> > (this assumes the z dimension of the box doesn't change, e.g. semiisotropic
> > pressure coupling with compressibility of zero along z).  Then supply this
> > file to grompp -r with a suitable [position_restraints] section that
> > specifies a flat-bottom potential along z (see manual section 4.3.2, I
> > think this is a case where you need a negative force constant to keep the
> > atoms "outside" of the restraint region). During the run, if any protein
> > atom hits the imaginary wall, it gets pushed back in the other direction by
> > the flat-bottom restraint.
> >
> > Also, don't couple solvent and ions to separate thermostats.  It's not
> > sensible and usually not stable.
> >
> > -Justin
> >
> > --
> > ==================================================
> >
> > 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.
> >
>
>
> ------------------------------
>
> Message: 5
> Date: Sun, 3 May 2015 11:57:56 -0400
> From: shivangi nangia <shivangi.nangia at gmail.com>
> To: Discussion list for GROMACS users <gmx-users at gromacs.org>,        Tsjerk
>       Wassenaar <tsjerkw at gmail.com>
> Subject: Re: [gmx-users] Secondary Structure Restraints: martinize.py
> Message-ID:
>       <CAH137_+=1V94ZHQp2LaSqvd84Oia169S_SGJT+8OeUFLDHN50w at mail.gmail.com>
> Content-Type: text/plain; charset=UTF-8
>
> Hello Tsjerk,
>
>
> I am executing the following command to get half helical and half random cg
> structure
>
> The pdb I am supplying is a helical peptide
>
> ./martinize.py -f gamma1_unmutated.pdb -x gamma1_unmutated.elnedyn.pdb -o
> gamma1.elnedyn.top  -ss LLLLLLLLLLHHHHHHHHHHH -p backbone  -ff martini22 >
>  martinize.out
>
> Then I minimize and rebox this cg peptide and reverse cg
>
> ./my_intiram.sh -f ../elnedynoff/reboxed_min_gamma1.gro -p protein_aa.top
> -po prot_bkmapped.top -from martini -to charmm36
>
> The backmapped.gro I get is not half random/half helical.
>
> Is there anything that I am doing wrong?
>
> Kindly guide.
>
> Thanks,
> sxn
>
> On Fri, May 1, 2015 at 4:43 PM, Tsjerk Wassenaar <tsjerkw at gmail.com> wrote:
>
> > Hi sxn,
> >
> > With martinize.py -ss LLLLLLLLLLLHHHHHHHHHH
> > you'll get the secondary structure you want.
> >
> > Cheers,
> >
> > Tsjerk
> >
> > On Fri, May 1, 2015 at 10:24 PM, shivangi nangia <
> > shivangi.nangia at gmail.com>
> > wrote:
> >
> > > Hello Tsjerk,
> > >
> > > Thanks for the reply.
> > >
> > > I have a 21 amino acids helical peptide, I want to first 10 amino acids
> > to
> > > not having secondary structure restraints while coarse graining using
> > > martinize.py and the rest to remain helical.
> > >
> > > I reverse coarse the peptide to see if my trials while playing around
> > with
> > > different trials with secondary structure file (I provide with -ss)
> > changes
> > > worked or not. For example, in one of my trials I changed the structure
> > > column of the .dssp file from H to nothing hoping once I coarse grain
> > with
> > > that modified file I would get a semi secondary structure restrained
> > coarse
> > > grained peptide.
> > >
> > > Thanks,
> > >
> > > sxn
> > >
> > > On Fri, May 1, 2015 at 3:59 PM, Tsjerk Wassenaar <tsjerkw at gmail.com>
> > > wrote:
> > >
> > > > Hi sxn,
> > > >
> > > > What do you mean with reverse coarse graining?
> > > > You can specify the secondary structure also as string, which is doable
> > > if
> > > > the peptide is not too long, e.g. using -ss HHHHHHHLLLLLLL. Otherwise,
> > > you
> > > > can put the secondary structure you want in a simple file, and pass
> > that
> > > > with the option -ss.
> > > >
> > > > Cheers,
> > > >
> > > > Tsjerk
> > > >
> > > >
> > > > On Fri, May 1, 2015 at 4:24 PM, shivangi nangia <
> > > shivangi.nangia at gmail.com
> > > > >
> > > > wrote:
> > > >
> > > > > Hello,
> > > > >
> > > > > I wish to turn off the secondary structure restraint in half of the
> > > > peptide
> > > > > while using martinze.py.
> > > > >
> > > > > Is there any way in which this can be done.
> > > > >
> > > > > I have tried to delete the columns in .dssp file which tell about the
> > > > > structure and character of the peptide but I still get a completely
> > > > helical
> > > > > peptide on reverse coarse graining.
> > > > >
> > > > > Thanks in advance.
> > > > >
> > > > > Best,
> > > > > sxn
> > > > > --
> > > > > 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.
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Tsjerk A. Wassenaar, Ph.D.
> > > > --
> > > > 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.
> > > >
> > > --
> > > 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.
> > >
> >
> >
> >
> > --
> > Tsjerk A. Wassenaar, Ph.D.
> > --
> > 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.
> >
>
>
> ------------------------------
>
> --
> 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.
>
> End of gromacs.org_gmx-users Digest, Vol 133, Issue 9
> *****************************************************

                                    mahender has files to share with you on OneDrive. To view them, click the links below.
--
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.


More information about the gromacs.org_gmx-users mailing list