[gmx-users] creating an "infinite" filament with editconf

[David van der Spoel]

On 07/12/15 14:06, Igaev, Maxim wrote:
> Thanks David, thanks Vitaly for your answers.
>
> @Vitaly: I also thought of this but the number of boundary atoms is too high. I'm afraid of introducing too many artifacts through the freezing.
>
> @David: the default -dd option of mdrun says that gromacs will guess the optimal size of the domains. The same for -npme. Do you mean I should just vary the size of the domains to find an optimal decomposition grid? I rather thought that the parts of my dimer that are not in the box after editconf and genbox should be "put"/"wraped" back into the box somehow...
>
>
They do. The error you show is an implementation limitation that you 
have to learn how to work with. Try -npme 0 to start with.
>
> ________________________________________
> Von: gromacs.org_gmx-users-bounces at maillist.sys.kth.se [gromacs.org_gmx-users-bounces at maillist.sys.kth.se]" im Auftrag von "David van der Spoel [spoel at xray.bmc.uu.se]
> Gesendet: Montag, 7. Dezember 2015 12:26
> An: gmx-users at gromacs.org
> Betreff: Re: [gmx-users] creating an "infinite" filament with editconf
>
> On 07/12/15 11:54, Igaev, Maxim wrote:
>> Dear Gromacs Users,
>>
>> I have a protein dimer and would like to make an "infinite" filament out of it in z direction by using periodic boundary conditions and editconf. What I've done so far is
>>
>> a) I've generated a topology for the dimer via
>>
>> pdb2gmx -f dimer.pdb -o dim.gro -p dim.top -i dim.itp -water tip3p    (force field was chosen interactively)
>>
>> b) and created a periodic box by using editconf
>>
>> editconf -f dim.gro -o dim_box.gro -bt triclinic -box 13.4 9.4 8.6 -center 6.7 4.7 4.3
>>
>>
>> If I understand editconf correctly, it is supposed to create a periodic box of sizes 13.4 x 9.4 x 8.6 and place the dimer's center of mass to (6.7, 4.7, 4.3) in this box. The size "8.6 nm" in z direction is the priodicity of my dimer, which means that the dimer doesn't fit completely into the box; some parts are sticking out. But this is needed to establish the periodicity.
>>
>> When I try to equilibrate my system after sufficient energy minimization I get an error like "X particles communicated to PME node Y are more than a cell length out of the domain decomposition cell of their charge group", which means that my system is blowing up and I probably have a bad starting structure. I then checked whether my dimer has some steric clashes and it was fine. I could equilibrate my dimer in a normal water box (isolated dimer) and the simulation was just fine.
>>
> You're doing everything right, but will have to tweak the mdrun command
> line settings to control the parallellization, in particular -dd and -npme
>> I wonder if I did something wrong during the editconf step. Should I wrap my dimer into the box after editconf so that I have no parts sticking out of the box?
>> In NAMD, a similar approach worked well - I just took my dimer and surrounded it by waters to have a box of sizes 13.4 x 9.4 x 8.6 and applied PBC to it. But how to do the same in GROMACS?
>>
>>
>> Thank you in advance for your help!
>>
>> Cheers,
>> Maxim
>>
>
>
> --
> David van der Spoel, Ph.D., Professor of Biology
> Dept. of Cell & Molec. Biol., Uppsala University.
> Box 596, 75124 Uppsala, Sweden. Phone:  +46184714205.
> spoel at xray.bmc.uu.se    http://folding.bmc.uu.se
> --
> 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.
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se