[gmx-users] Simulating conjugated protein

[Justin Lemkul]

On 6/2/16 8:14 AM, Sanket Ghawali wrote:
>>
>>
>> Date: Wed, 25 May 2016 07:11:25 -0400
>> From: Justin Lemkul <jalemkul at vt.edu>
>> To: gmx-users at gromacs.org
>> Subject: Re: [gmx-users] Simulating conjugated protein
>> Message-ID: <e0dc7a46-bfcd-ebb0-c188-dd993682a0d8 at vt.edu>
>> Content-Type: text/plain; charset=windows-1252; format=flowed
>>
>>
>>
>> On 5/25/16 12:55 AM, Sanket Ghawali wrote:
>>> Dear, gmx-users,
>>>
>>> Hello everyone,
>>>
>>> I'm simulating a peptide in an SDS micelle, where my peptide is
>> conjugated
>>> to a organic compound at the N terminal. The SDS micelle was genrated
>> using
>>> CHARMM-GUI, here I am stuck with a problem for adding atom types for the
>>> compound. I tried adding atomtypes for the compound into the .rtp files
>> of
>>> the CHARMM forcefield but when i try to run pdb2gmx it throws an error
>>> regarding missing residues in .hdb file.
>>>
>>> I edited the .hdb file adding the hydrogen atom information for the
>> unknown
>>> compound. The pdb2gmx worked this time but it throws back an error while
>>> adding ions.
>>>
>>> I tried using SwissParam to parameterized the unknown molecule, using
>> this
>>> i find that the compound does not remain bound to the peptide at the N
>>> terminal even after merging the compound and peptide co ordinates.
>>>
>>> The errors which i get while adding ions:
>>>
>>> ERROR 190 [file topol.top, line 3284]:
>>>   No default Proper Dih. types
>>>
>>>
>>> ERROR 191 [file topol.top, line 3285]:
>>>   No default Proper Dih. types
>>>
>>>
>>> ERROR 192 [file topol.top, line 3286]:
>>>   No default Proper Dih. types
>>>
>>> Excluding 3 bonded neighbours molecule type 'Protein'
>>> Excluding 3 bonded neighbours molecule type 'SDS'
>>> Excluding 2 bonded neighbours molecule type 'SOL'
>>> Excluding 1 bonded neighbours molecule type 'NA'
>>>
>>> NOTE 1 [file topol.top, line 3377]:
>>>   System has non-zero total charge: 17.180007
>>>   Total charge should normally be an integer. See
>>>   http://www.gromacs.org/Documentation/Floating_Point_Arithmetic
>>>   for discussion on how close it should be to an integer.
>>>
>>>
>>>
>>>
>>> -------------------------------------------------------
>>> Program grompp, VERSION 4.6.5
>>> Source code file: /root/Downloads/gromacs-4.6.5/src/kernel/grompp.c,
>> line: 563
>>>
>>> Fatal error:
>>> number of coordinates in coordinate file (sol.gro, 15465)
>>>              does not match topology (topol.top, 15450)
>>> For more information and tips for troubleshooting, please check the
>> GROMACS
>>> website at http://www.gromacs.org/Documentation/Errors
>>> -------------------------------------------------------
>>>
>>> "Being Great is Not So Good" (Red Hot Chili Peppers)
>>>
>>> .
>>>
>>>
>>> Any suggestions on how to go about doing this?
>>>
>>
>>
>> http://www.gromacs.org/Documentation/How-tos/Adding_a_Residue_to_a_Force_Field#Adding_a_new_residue
>>
>> You need to properly parametrize the residue in question first, including
>> the
>> whole linkage.  Based on the missing parameters and the suspicious charge,
>> your
>> underlying parametrization is not functional.
>>
>> -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
>>
>> ==================================================
>> Thank you Justin for ur reply
>>
>> I am using Swissparam and Cgenff but the only work for small molecules. I
>> was able to generate it for the compound but my peptide is covalently
>> linked to the compound at the N terminal so how do i parametrize the whole
>> linkage or the whole compound.

[compound]-NMA

If the assigned charges on the NMA portion of the molecule differ from the 
protein backbone, adjust the charges on the linkage of the remainder of the 
compound to compensate.  You want to merge the new molecule as smoothly as 
possible with existing parameters.

Ideally you will not use CGenFF types for this; use the protein force field as a 
basis if there is going to be a covalent bond.  You can use CGenFF to get 
initial charges and parameters for unknown bonded interactions, but you should 
thoroughly refine that initial topology.  Linking something with an already 
highly optimized force field requires a similar level of care.  Garbage in, 
garbage out.

>> And what name should be given to the compound in the residuetype.dat file.

Whatever you want it to be.

>> Following the SwissParam tutorial:
>> http://www.swissparam.ch/
>> i was able to work it out upto energy minimization step, I saved the
>> initial coordinate of the peptide from the .gro file into .pdb file using
>> VMD, when i visualize the .pdb file Discovery studio the compound is not
>> bound at the N terminal.

Visualization is not definitive.  The topology is.  Either there is a bond there 
or there isn't.  It can't break during EM/MD, so either the topology is right or 
it is wrong.

>> The tutorial i am using is it correct ?

No clue.  I don't use SwissParam.

-Justin

>> Guide me if i am going wrong somewhere.
>>
>
>     Thank you.
>

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

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

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