Hello Thomas,
I hope you are doing well !
I have a small question regarding the protocol of the parametrization of
the system with two identical ligands. Here are three possibilities for the
parametrization:
(1) if I specify the two replicas of the same ligands in the initial pdb
as "lig" and "lih" and subsequently grep each unit from the complex for the
parametrization using antechamber (so totally I run it two times, which
could not be correct according to your previous message) everything works
fine. In the protein.prmtop I have two units for lig and lih and it looks
OK.
(2) if I try to name the both units as "lig" and parametrize only one of
them and then apply new parameters on the both replicas (in the initial
pdb) it works also OK. In the parmtop file I have two identical units: lig,
lig. But as I talked before the presence of two identical units would
complicate analysis.
(3) Finally in the case when I still keep in the model two different names
"lig" and "lih" but run parametrization only one time (e.g. using only
"lig"), obviously parmtop for the complex could not be created and tleap
sent an error message that all atoms of "lih" did not have parameters.
How could I fix the (3) in order that I could have two different names for
the same unit in the model ? Or alternatively should I rename one of the
"lig" in the case of the (2) after its parametrization ?
Many thanks in advance !
Enrico
Il giorno ven 24 mar 2023 alle ore 10:11 Steinbrecher, Thomas <
thomas.steinbrecher.roche.com> ha scritto:
> Hi Enrico,
>
> Im not sure parameterizing twice is a good idea, your parameters may be
> slightly different due to input structure bias, so you have two subtly
> different 'identical' ligands in your system. What's wrong with
> parameterizing the molecule once and using those parameters for both
> residues? You can specify them later with e.g. ":lig,lih" but I'd really
> recommend the ambmask chapter 23.1.1 of the Amber manual for more details
> and examples, masks can be tricky to write and you need to make sure they
> match what you really want to be selected, especially if you are aiming for
> some semi-automated analysis later.
>
> Are you sure you want both molecules as the "ligand" in MMPBSA? Then you'd
> be neglecting entropy effects twice leading to scores that are hard to
> interpret. I'm not saying this is wrong, depending on what you aim for, but
> it sounds like you are planning some pretty non-standard MMPBSA
> calculations here.
>
> Kind Regards,
>
> Thomas
>
> On Thu, Mar 23, 2023 at 5:35 PM Enrico Martinez <jmsstarlight.gmail.com>
> wrote:
>
>> Okay, thank you very much Thomas!
>>
>> To introduce more flexibility in the analysis routine, I decided to use
>> two different names for identical molecules and reparametrize each twisely
>> via antechamber using identical parameters .. So imagine that if I have
>> :lig and :lih what should be amber mask to use with anti-MMPBSA.py to
>> check and select the both of them at once as a ligand ? :-)
>>
>> Many thanks in advance
>>
>> Enrico
>>
>> Il giorno gio 23 mar 2023 alle ore 15:53 Steinbrecher, Thomas <
>> thomas.steinbrecher.roche.com> ha scritto:
>>
>>> Hi Enrico,
>>>
>>> identical molecules need only be parameterized once. So you would use
>>> antechamber (and parmchk/parmfit ?) to create ligand.mol2 (or .lib) files
>>> describing your compound. Normally, identical molecules would then have the
>>> same residue name in your system, but that is not mandatory, just
>>> convention. So in your case, when building your system from a pdb file, you
>>> need to tell leap what parameters to use for each residue name, so if your
>>> ligands are called LIG and LI2, then these two entries need to be defined
>>> in your leap input file, for example by the two commands I posted above.
>>>
>>> Kind Regards,
>>>
>>> Thomas
>>>
>>>
>>>
>>> On Thu, Mar 23, 2023 at 3:27 PM Enrico Martinez via AMBER <
>>> amber.ambermd.org> wrote:
>>>
>>>> Okay, thank you !
>>>>
>>>> Regarding the renaming, should it be in principle possible to
>>>> parametrize
>>>> the complex with two identical residues having different names in the
>>>> initial complex e.g. lig1 and lig2 or for such case I need to
>>>> parametrize
>>>> each ligand separately using antechamber ?
>>>>
>>>> BTW, I've just executed mmgbsa using :lig mask and it seems it's
>>>> summarized
>>>> all the values resulting from from the calculations for both residues,
>>>> so
>>>> .. :- )
>>>>
>>>> Enrico
>>>>
>>>> Il giorno gio 23 mar 2023 alle ore 15:08 Carlos Simmerling <
>>>> carlos.simmerling.gmail.com> ha scritto:
>>>>
>>>> > but if you want it for mmgbsa then this won't work... maybe renaming
>>>> the
>>>> > second ligand is better
>>>> >
>>>> > On Thu, Mar 23, 2023 at 10:06 AM Carlos Simmerling <
>>>> > carlos.simmerling.gmail.com> wrote:
>>>> >
>>>> >> you can try making a for loop in cpptraj, you can loop over all
>>>> >> occurrences of the ligand and write the analysis separately for each
>>>> ligand
>>>> >> instance, then only use the data file that you want. you could also
>>>> modify
>>>> >> the script to only analyze the first occurrence, but that would be a
>>>> little
>>>> >> more work in the script.
>>>> >>
>>>> >> try something like this, where you find matching residues, and then
>>>> write
>>>> >> the outputs using the residue number so you know later which is
>>>> first.:
>>>> >>
>>>> >> parm prmtop.parm7
>>>> >> trajin mdcrd.nc
>>>> >> for residues A0 inmask :LIG
>>>> >> rms \$A0 out rms.\$A0.dat nofit
>>>> >> done
>>>> >>
>>>> >>
>>>> >> On Thu, Mar 23, 2023 at 9:58 AM Enrico Martinez <
>>>> jmsstarlight.gmail.com>
>>>> >> wrote:
>>>> >>
>>>> >>> Thanks a lot, Carlos !
>>>> >>>
>>>> >>> Are there any other possibilities to select the ligand without
>>>> residue
>>>> >>> number ?
>>>> >>>
>>>> >>> E.g. I have two systems with two different proteins bound twisely
>>>> to the
>>>> >>> same ligand in chain A and chain B.
>>>> >>> I need to use an atom mask, which would select the first occurrence
>>>> of
>>>> >>> the residue lig in the complex w/o specification of the residue
>>>> number
>>>> >>> which could differ in various systems..
>>>> >>>
>>>> >>> Many thanks in advance
>>>> >>>
>>>> >>> Enrico
>>>> >>>
>>>> >>> Il giorno gio 23 mar 2023 alle ore 13:19 Carlos Simmerling <
>>>> >>> carlos.simmerling.gmail.com> ha scritto:
>>>> >>>
>>>> >>>> You can also use the residue number.
>>>> >>>>
>>>> >>>> On Thu, Mar 23, 2023, 8:17 AM Enrico Martinez via AMBER <
>>>> >>>> amber.ambermd.org> wrote:
>>>> >>>>
>>>> >>>>> Dear Friends,
>>>> >>>>>
>>>> >>>>> Thank you very much for your kind suggestions!
>>>> >>>>> Indeed this was an issue in my preparation setup using
>>>> antechamber,
>>>> >>>>> which
>>>> >>>>> took both residues at the same time.
>>>> >>>>>
>>>> >>>>> BTW, I have a question related to the analysis of the system
>>>> >>>>> containing two
>>>> >>>>> identical ligands (the both named as LIG) bound in two different
>>>> >>>>> monomers
>>>> >>>>> of the complex. How could it be better to specify the atom mask
>>>> >>>>> corresponding to a specific ligand (e.g. for mmgbsa) ? The only
>>>> one
>>>> >>>>> solution could be based only on the residue name in this case ?
>>>> >>>>>
>>>> >>>>> Many thanks in advance !
>>>> >>>>>
>>>> >>>>> Yours sincerely
>>>> >>>>>
>>>> >>>>> Enrico
>>>> >>>>>
>>>> >>>>> Il giorno mar 21 mar 2023 alle ore 17:27 Dr. Anselm Horn via
>>>> AMBER <
>>>> >>>>> amber.ambermd.org> ha scritto:
>>>> >>>>>
>>>> >>>>> > Dear Enrico,
>>>> >>>>> >
>>>> >>>>> > it is not clear to me what you are doing:
>>>> >>>>> > If you already have the parameters for your ligand, then you
>>>> need to
>>>> >>>>> > load them only once into leap; the programs identifies the
>>>> ligand
>>>> >>>>> > according to its residue name, and thus it can deal with several
>>>> >>>>> > identical residues in your (pdb) input, as long as the residue
>>>> names
>>>> >>>>> in
>>>> >>>>> > your input and the parameter files do match.
>>>> >>>>> >
>>>> >>>>> > If you are still in the parameterization phase, then you also
>>>> need
>>>> >>>>> only
>>>> >>>>> > one instance of your ligand.
>>>> >>>>> >
>>>> >>>>> > Best,
>>>> >>>>> >
>>>> >>>>> > Anselm
>>>> >>>>> >
>>>> >>>>> > Bioinformatik | NHR.FAU
>>>> >>>>> > Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU)
>>>> >>>>> > Germany
>>>> >>>>> >
>>>> >>>>> >
>>>> >>>>> > Am 21.03.2023 um 17:00 schrieb Enrico Martinez via AMBER:
>>>> >>>>> > > Dear Amber Users !
>>>> >>>>> > >
>>>> >>>>> > > I have an issue related to the preparation of the system
>>>> >>>>> containing two
>>>> >>>>> > > identical ligands present in two different binding sites.
>>>> Actually
>>>> >>>>> when I
>>>> >>>>> > > try to copy the both units for the parametrization using
>>>> >>>>> Antechamber it
>>>> >>>>> > > says that it could not process two residues at the same time.
>>>> >>>>> > >
>>>> >>>>> > > Is there any preparation trick for such a case assuming that
>>>> >>>>> everything
>>>> >>>>> > > works fine with the same system containing only one ligand ?
>>>> >>>>> > >
>>>> >>>>> > > Many thanks in advance
>>>> >>>>> > >
>>>> >>>>> > > Enrico
>>>> >>>>> > > _______________________________________________
>>>> >>>>> > > AMBER mailing list
>>>> >>>>> > > AMBER.ambermd.org
>>>> >>>>> > > http://lists.ambermd.org/mailman/listinfo/amber
>>>> >>>>> > >
>>>> >>>>> >
>>>> >>>>> >
>>>> >>>>> > _______________________________________________
>>>> >>>>> > AMBER mailing list
>>>> >>>>> > AMBER.ambermd.org
>>>> >>>>> > http://lists.ambermd.org/mailman/listinfo/amber
>>>> >>>>> >
>>>> >>>>> _______________________________________________
>>>> >>>>> AMBER mailing list
>>>> >>>>> AMBER.ambermd.org
>>>> >>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>> >>>>>
>>>> >>>>
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>
>>>
>>>
>>> --
>>> *Dr. Thomas Steinbrecher*
>>> Principal Scientist CADD
>>>
>>> Roche Pharma Research and Early Development
>>> Roche Innovation Center Basel
>>> F. Hoffmann-La Roche Ltd
>>> Bldg. 092/3.92
>>> Grenzacherstrasse 124
>>> 4070 Basel
>>> Switzerland
>>>
>>> Phone +41 61 682 1319
>>> mailto: thomas.steinbrecher.roche.com
>>>
>>
>
> --
> *Dr. Thomas Steinbrecher*
> Principal Scientist CADD
>
> Roche Pharma Research and Early Development
> Roche Innovation Center Basel
> F. Hoffmann-La Roche Ltd
> Bldg. 092/3.92
> Grenzacherstrasse 124
> 4070 Basel
> Switzerland
>
> Phone +41 61 682 1319
> mailto: thomas.steinbrecher.roche.com
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Mar 27 2023 - 07:00:02 PDT
