[gmx-users] on md of protein-ligand

Wed Apr 20 11:49:21 CEST 2016

Hi,

On Wed, Apr 20, 2016 at 11:21 AM Brett <brettliu123 at 163.com> wrote:

> Hi Mark,
>
>
>
> I still find it was difficult to understand. After the pdb2gmx, the
> coordinates in the gro file will be significantly (including the scale of
> the value) different from the input for pdb2gmx. Suppose my protein-ligand
> contains 6 identical ligands, the ATB like servers can only produce one set
> of topology and gro files for all 6 identical ligands (with 6 different
> initial pdbs).
>

OK, but probably the output coordinates are useless.

> Then how do we get a single gro file containing all the coordinates
> information for the 6 subunits (relatively easy for the protein part by
> pdb2gmx) and all 6 ligands (as for we can only hav 1 set of coordinates
> information containing ligand.gro for all 6 ligands)?
>

I still don't know what your problem is. You don't need a file in .gro
format. That just happens to be the default output format of pdb2gmx. You
need an initial configuration, in any format with your ligands placed
wherever you want them - this will require some creative use of editconf,
or some non-GROMACS structure-building tool. You will need a topology file
whose molecule ordering is the same as the configuration file, and whose
atom ordering within molecules also matches. And then you'll need to
solvate.

Mark

> Brett
>
>
>
>
>
>
>
>
>
> At 2016-04-20 16:56:09, "Mark Abraham" <mark.j.abraham at gmail.com> wrote:
> >Hi,
> >
> >The topology and the coordinates are two separate sets of infomation. You
> >already expect that lot of conformations will be consistent with the
> >topology, and explored over the simulations. The topology generating
> >servers want a structure so they can make good chemical guesses for you.
> >But later you can combine the topology with any conformation, or even use
> >the same one for multiple identical molecules. Whether the resulting model
> >later produces good results depends on the quality of the parametrisation
> >of the whole system.
> >
> >Mark
> >
> >On Wed, 20 Apr 2016 09:47 Brett <brettliu123 at 163.com> wrote:
> >
> >> Dear All and Justin,
> >>
> >>
> >> Suppose my protein complex contain 4 ADP, each ADP was in one subunit.
> >> Suppose we use ATB to create the ligand ADP files for md. It seems the
> >> situation is, regardless of which of the 4 ADP pdb files was submitted,
> ATB
> >> will only give one identical set of topology files for ADP, and it
> cannot
> >> distinguish from which ADP ligand ATB  got the topology files for all 4
> >> ADPs.
> >>
> >>
> >> Here it lead to the question, how does GROMCS knows the position of each
> >> ADP in the protein complex during the md process?
> >>
> >>
> >> Let us think the question from another view. We have a protein-ligand
> >> complex pdb, and the complex contains 1 subunit and 1 ligand. We got the
> >> protein topology file by pdb2gmx, and we will find in the gro file the
> >> value of the coordinates have been changed from the original pdb
> >> coordinates for the protein-ligand. The same is true for the ligand,
> and if
> >> we use prodrug to get the topology file for the ligand, we will find the
> >> coordinate values of the ligand have been changed in comparison with
> that
> >> in the ligand pdb from the protein-ligand pdb.
> >>
> >>
> >> Thus, I would like to ask, after we edited the gro files by combining
> the
> >> protein part coordinates and ligand part coordinates, how GROMCS keeps
> that
> >> once the compiled gro file was transformed back to pdb, the ligands
> were in
> >> the correct ligand binding pocket in the protein?
> >>
> >> I am looking forward to getting the reply from you.
> >>
> >>
> >> Brett
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >> At 2016-04-19 20:13:03, "Justin Lemkul" <jalemkul at vt.edu> wrote:
> >> >
> >> >
> >> >On 4/19/16 8:10 AM, Brett wrote:
> >> >> Dear All,
> >> >>
> >> >>
> >> >> It seems the external servers preparing the ligand files (antechamber
> >> for example) not only optimize the coordinates of the ligands, but it
> >> changes the ligand from one place to another place, thus the ligand
> >> coordinates by the external server cannot occupy the ligand binding
> pocket
> >> in the original protein-ligand complex.
> >> >>
> >> >>
> >> >> I am looking forward to getting a reply from you on how to have the
> >> external server processed ligand find the ligand binding pocket in the
> >> protein-ligand complex for md.
> >> >>
> >> >
> >> >Topology-generating servers aren't going to "find the ligand binding
> >> pocket" -
> >> >your job is to make sure the original coordinates are preserved, as the
> >> previous
> >> >message has instructed you on how to do.
> >> >
> >> >-Justin
> >> >
> >> >>
> >> >> Brett
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >> At 2016-04-19 19:44:23, "suniba" <sun.iba2 at gmail.com> wrote:
> >> >>> After pdb2gmx, protein topology is prepared. You prepare the ligand
> >> topology from an external server or parametrize it. Then you prepare
> >> complex.gro and include the ligand topology manually in topol.top
> (topology
> >> generated by pdb2gmx). And for your second question, I used ATB today
> and
> >> realized that one should always use "original geometry" co-ordinates as
> >> 'optimized geometry' will/may cause clashes with protein etc.. So, avoid
> >> PRODRG server and use ATB and then download the co-ordinates and .itp of
> >> original geometry. Rest process is same as mentioned in tutorial.
> >> >>> Regards
> >> >>>
> >> >>> Sent from my iPhone
> >> >>>
> >> >>>> On 19-Apr-2016, at 4:58 pm, Brett <brettliu123 at 163.com> wrote:
> >> >>>>
> >> >>>> Dear All,
> >> >>>>
> >> >>>>
> >> >>>> I am learning the md of protein-ligand based on the Justin on-line
> >> tutorial. My first question is, for the topology files we need to
> produce
> >> the protein part and ligand part separately, and the input for pdb2gmx
> did
> >> not contain the ligand part. After I got the ligand files, I find the
> >> coordinate of the ligands was different from the ligand pdb in the
> >> protein-complex pdb (although the rmsd between the ligand topology and
> the
> >> ligand in the complex was 0). Then during md process how does GROMCS
> know
> >> where is the position of the ligand in the complex?
> >> >>>>
> >> >>>>
> >> >>>> Correspondingly, my second question is, suppose I have a protein
> >> dimer with 2 identical subunits, 1 subunit was a cAMP binding, and 1 was
> >> apo. Then during the md process how does GROMACS know which subunit
> binding
> >> with the cAMP and which subunit was apo?
> >> >>>>
> >> >>>>
> >> >>>> I am looking forward to getting a reply from you.
> >> >>>>
> >> >>>>
> >> >>>> Brett
> >> >>>> --
> >> >>>> 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.
> >> >
> >> >--
> >> >==================================================
> >> >
> >> >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.
> >> --
> >> 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.
> --
> 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.
>