[gmx-users] problem with gmx genconf?

Justin Lemkul jalemkul at vt.edu
Mon Oct 10 03:33:21 CEST 2016



On 10/9/16 9:22 PM, Thejus Kartha wrote:
> Well, the literature reports a non-zero dipole moment and hence, a polarity index of 4.8 for 1,4-dioxane. So does the organic chemist nearby. I'm not sure if I understand your perspective, Dr. Lemkul. I'm clearly missing something here.
>

It's a symmetric molecule, so if there is a net dipole moment, it should be very 
small, if any at all (some tables list the value of 0.45 D).  Note that an in 
vacuo dipole moment will differ from a condensed phase value (because there is a 
surrounding electric field in the condensed phase).

Of course there will be small variations depending on geometry.  You should do a 
QM optimization of the structure and compute its dipole moment, then do the same 
calculations as a function of restrained geometries (planar, chair, boat, twist, 
etc).  This would serve as suitable target data for deriving charges.

> I should also mention that I have run simulations of ethanol which has a polarity index of 5.2 on the same scale as 1,4-dioxane. These were done on a GROMOS43a1 forcefield too, on GROMACS. This leaves me thoroughly confused as what to do.
>

I don't see how this is related.

> Also, if I am to model such polar solvents in the future, is there a particular force field which you could suggest?
>

With proper parameters, you can use any force field you like.

-Justin

>
>     On Monday, 10 October 2016 6:14 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
>
>
>
> On 10/9/16 8:26 PM, Thejus Kartha wrote:
>> I'm sorry Dr. Lemkul. I thought I put the links for the images I uploaded on Google Drive. Let me try it again.
>> The single molecule:
>> https://drive.google.com/open?id=0B3dugKVFN2ajTmZMYVdOX0NfbGM
>> The system that was generated by genconf:https://drive.google.com/open?id=0B3dugKVFN2ajSDdiampwREVoeUU
>>
>
> Don't use -shuf.  You're scrambling your molecules.
>
>> Regarding the charges in the topology, I produced them myself using a minimization in Avogadro. I was guessing that was sufficient. It also made sense to me, since dioxane is a polar molecule with non-zero dipole moment. There is no necessity that the system I deal with must have zero charge, is there?
>>
>
> This is wrong.  It is a symmetrical molecule and in the absence of an external
> electric field, will in fact have a zero net dipole moment.  It may have
> non-zero components of the dipole vector, but the total net dipole is zero.
>
> The fractional charges absolutely must sum to zero in this molecule.  Ionized
> molecules can have *integral* charges (-1, +1, etc) but no molecule can have a
> fractional charge.  You can't acquire 46.6% of an electron, can you?  That's
> what your topology is doing.
>
> You should be building this topology based on any available linear or cyclic
> ether parameters available in GROMOS, or use a force field that you can
> parametrize more easily.
>
> -Justin
>
>> Thank you.
>> Thejus Kartha
>>
>>     On Monday, 10 October 2016 5:47 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>>
>>
>>
>> None of your screenshots work; most users receive mail in plain text.  If you
>> want to share files or images, upload them somewhere and provide a link.
>>
>> <snip>
>>
>>> [ atoms ]
>>> ;  nr      type  resnr residue  atom  cgnr    charge      mass  typeB    chargeB      massB
>>> ; residue 903 DOX rtp DOX  q -0.5
>>>       1        CH2    903    DOX    CA      1      0.071    14.027  ; qtot 0.071
>>>       2        CH2    903    DOX    CB      1      0.071    14.027  ; qtot 0.142
>>>       3        CH2    903    DOX    CD      1      0.071    14.027  ; qtot 0.213
>>>       4        CH2    903    DOX    CE      1      0.071    14.027  ; qtot 0.284
>>>       5        OA    903    DOX    OF      1    -0.375    15.9994  ; qtot -0.091
>>>       6        OA    903    DOX    OG      1    -0.375    15.9994  ; qtot -0.466
>>
>> This topology is complete nonsense.  You have a net charge of -0.466.  I don't
>> know where you got those charges, but they make no sense whatsoever.
>>
>> -Justin
>>
>>>
>>> [ bonds ]
>>> ;  ai    aj funct            c0            c1            c2            c3
>>>     1    2    2    gb_14
>>>     1    6    2    gb_12
>>>     2    5    2    gb_12
>>>     3    4    2    gb_14
>>>     3    5    2    gb_12
>>>     4    6    2    gb_12
>>>
>>> [ pairs ]
>>> ;  ai    aj funct            c0            c1            c2            c3
>>>     1    3    1
>>>     2    4    1
>>>     5    6    1
>>>
>>> [ angles ]
>>> ;  ai    aj    ak funct            c0            c1            c2            c3
>>>     2    1    6    2    ga_14
>>>     1    2    5    2    ga_14
>>>     4    3    5    2    ga_14
>>>     3    4    6    2    ga_14
>>>     2    5    3    2    ga_9
>>>     1    6    4    2    ga_9
>>>
>>> [ dihedrals ]
>>> ;  ai    aj    ak    al funct            c0            c1            c2            c3            c4            c5
>>>     6    1    2    5    1    gd_1
>>>     2    1    6    4    1    gd_2
>>>     1    2    5    3    1    gd_2
>>>     5    3    4    6    1    gd_1
>>>     4    3    5    2    1    gd_2
>>>     3    4    6    1    1    gd_2
>>>
>>> ; Include Position restraint file
>>> #ifdef POSRES
>>> #include "posre.itp"
>>> #endif
>>>
>>> ; Include water topology
>>> #include "gromos43a1.ff/spce.itp"
>>>
>>> #ifdef POSRES_WATER
>>> ; Position restraint for each water oxygen
>>> [ position_restraints ]
>>> ;  i funct      fcx        fcy        fcz
>>>     1    1      1000      1000      1000
>>> #endif
>>>
>>> ; Include topology for ions
>>> #include "gromos43a1.ff/ions.itp"
>>>
>>> [ system ]
>>> ; Name
>>> DIOXANE
>>>
>>> [ molecules ]
>>> ; Compound        #mols
>>> DOX              1--------------------------------------------------------------------------------------------------------
>>> But, when I use the gmx genconf command to expand it into a collection of molecules (8 x 8 x 8 box), the result contains disordered molecules as seen in my VMD window. However, the .gro file produced after the genconf seems to be looking fine though. The genconf generated system looks like this:
>>> Screenshot from 2016-10-10 05-12-09.png
>>> As you can see, the molecules are not seen correctly, that is, there are many places where I have an oxygen in the place of carbon, vice-versa, only carbon containing rings, so on and so forth. Can you please let me know what's going on? I mean, there's no reason why gmx genconf shouldn't be just duplicating the molecule I give it.
>>>
>>> The command I used:
>>> gmx genconf -f dioxane.gro -o diox.gro -nbox 8 8 8 -dist 0.3  -shuffle -rot
>>>
>>>
>>> |
>>> |
>>> |
>>> |  |    |
>>>
>>>     |
>>>
>>>   |
>>> |
>>> |    |
>>> Screenshot from 2016-10-10 05-12-09.png
>>>     |  |
>>>
>>>   |
>>>
>>>   |
>>>
>>>
>>> Thanks & regards,Thejus Kartha
>>>
>>
>

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

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

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


More information about the gromacs.org_gmx-users mailing list