[gmx-users] Small molecule parametrization

[Justin Lemkul]

On 9/9/16 8:21 AM, sun.iba2 wrote:
> This is a small molecule with dicyanovinyl group substituted at para position
> on one of the phenyl ring. I suspect the charges are not proper because in
> ATB the group charges must be -1,0 or 1. For some groups charges are summed
> up to zero but this is not the case of all groups.
>

There is no such requirement if you run the simulation with PME, and charge 
groups are anyway ignored with the Verlet cutoff scheme in GROMACS.

Your molecule is very complex.  I doubt any existing biomolecular force field 
will deal with it well.  You'll need to do very careful parametrization of 
several suitable model compounds (not the whole molecule) and link them together.

-Justin

>
> Sent from Samsung Mobile
>
> <div>-------- Original message --------</div><div>From: Justin Lemkul
> <jalemkul at vt.edu> </div><div>Date:09/09/2016  5:18 AM  (GMT+05:30)
> </div><div>To: gmx-users at gromacs.org </div><div>Cc:  </div><div>Subject: Re:
> [gmx-users] Small molecule parametrization </div><div> </div>
>
> On 9/7/16 1:43 PM, Sun Iba wrote:
>> Hi I have parametreized my small molecule with the help of ATB server. I
>> have learnt that ATB provides GROMOS charges better than PRODRG. However, I
>> have serious doubts regarding the charges assigned to hydrophobic groups.
>> Also I need to convert the calculated charges into GROMOS 43a1 set (force
>> field I am using). I am pasting the .itp file produced by ATB:
>>
>> nr  type  resnr  resid  atom  cgnr  charge    mass    total_charge 1   CH3
>> 1    SCX1    C12    1    0.223  15.0350 2    OE    1    SCX1     O1    1
>> -0.275  15.9994     ; -0.052 3     C    1    SCX1     C3    2    0.476
>> 12.0110 4     C    1    SCX1     C2    2   -0.157  12.0110 5     C    1
>> SCX1     C4    2   -0.376  12.0110 6    HC    1    SCX1     H2    2
>> 0.172   1.0080     ;  0.115 7     C    1    SCX1     C6    3    0.095
>> 12.0110 8    HC    1    SCX1     H4    3    0.137   1.0080 9     C    1
>> SCX1     C9    3   -0.386  12.0110 10     C    1    SCX1    C13    3
>> 0.478  12.0110 11    NR    1    SCX1     N2    3   -0.525  14.0067     ;
>> -0.201 12     C    1    SCX1    C19    4    0.539  12.0110 13    NR    1
>> SCX1     N4    4   -0.539  14.0067     ;  0.000 14     C    1    SCX1
>> C1    5   -0.117  12.0110 15    HC    1    SCX1     H1    5    0.157
>> 1.0080 16     C    1    SCX1     C5    5   -0.150  12.0110 17    HC    1
>> SCX1     H3    5    0.131   1.0080     ;  0.021 18     C    1    SCX1
>> C7    6    0.097  12.0110 19     C    1    SCX1    C11    6    0.247
>> 12.0110 20    HC    1    SCX1     H6    6    0.079   1.0080 21    NR    1
>> SCX1     N1    6   -0.423  14.0067     ;  0.000 22    NT    1    SCX1
>> N3    7    0.171  14.0067 23     C    1    SCX1    C16    7    0.387
>> 12.0110 24     O    1    SCX1     O2    7   -0.492  15.9994 25     C    1
>> SCX1    C18    7    0.495  12.0110 26     F    1    SCX1     F1    7
>> -0.187  18.9984 27     F    1    SCX1     F2    7   -0.187  18.9984 28
>> F    1    SCX1     F3    7   -0.187  18.9984     ;  0.000 29     C    1
>> SCX1    C15    8   -0.211  12.0110 30     C    1    SCX1    C14    8
>> 0.388  12.0110 31   CH3    1    SCX1    C10    8   -0.060  15.0350     ;
>> 0.117 32     C    1    SCX1    C22    9   -0.444  12.0110 33    HC    1
>> SCX1    H11    9    0.187   1.0080 34     C    1    SCX1    C21    9
>> 0.309  12.0110 35   CH3    1    SCX1     C8    9   -0.052  15.0350     ;
>> 0.000 36     C    1    SCX1    C20   10   -0.340  12.0110 37    HC    1
>> SCX1    H10   10    0.226   1.0080 38     C    1    SCX1    C17   10
>> -0.026  12.0110 39    HC    1    SCX1     H8   10    0.140   1.0080     ;
>> 0.000
>>
>
> What is this molecule and why do you think the charges are inappropriate?
>
>>
>> Please have a look and help me understanding if the charges are too bad.
>> Also my Nitrogen atom is not aromatic. It is linked to C by triple bond
>> and to another N by double bond. Does GROMOS treat this type of nitrogn as
>> NT and NR?
>>
>
> N can't be linked to one atom via triple bond and another via double bond.
> That's five bonds.  Unless your N has expanded its valence into d-orbitals,
> that's not going to be possible.
>
> -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

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