[gmx-users] Using CHARMM36 in GROMACS to simulate polysaccharides

[Justin Lemkul]

On 5/23/17 2:16 PM, Mohammad Hassan Khatami wrote:
> Thank Justin!
> I finished adding the new “residues” in the .rtp file. pdb2gmx runs straight forward. However, while doing energy minimizations, I get several errors like the ones below, arise:
>
> ERROR 2 [file topol.top, line 396]:
>   No default U-B types
> and
> ERROR 5 [file topol.top, line 550]:
>   No default Proper Dih. types
>
> As it says, they might be related to the angles and dihedral parameters. Do you have any suggestion on where (and how) to add them?

You shouldn't need to add parameters, but without knowing what is on those lines 
(more specifically, the associated atom types because the atom numbers will mean 
nothing to me) there's nothing else I can say about it.  The entirety of the 
carbohydrate force field should already be ported over, so missing parameters 
are very odd.

-Justin

> Thanks again,
> MH
>>>
>>> On 5/22/17 1:09 AM, Mohammad Hassan Khatami wrote:
>>>> Hi Justin,
>>>> I am asked to focus on learning how to change and update the CHARMM36 parameters, so I could implement the future changes and patches easier. (Thus, I am not focused in the Glycan Reader, at the moment.)
>>>>
>>>> Thank you! I think I now have a better understanding of what should I do. For each part of my polymer,i.e. the initial, the middle and the final part, I have to modify the AGLC molecule to represent each of these parts, seperately.
>>>> So, lets say to introduce the 1->4 linkage, I need to to apply 14ba patch from the top_all36_carb.rtf into the merged.rtp file of GROMACS CHARMM36. In this case, I need to create a new version of [ AGLC ] molecule (lets call it [ AGLC14 ]) in the merged.rtp, with the changes below from the the top_all36_carb.rtf, applied to it:
>>>
>>> Note that your residue name must be limited to 4 characters so it can properly be read from the input coordinates.  AGLC14 won't work.
>>>
>>>> ! equatorial-axial 1->4 linkage
>>>> PRES 14ba           0.02 ! (i)1->4(i-1) equatorial at C1 and axial at C4
>>>> dele atom 1HO4
>>>> dele atom 2HO1
>>>> dele atom 2O1
>>>> ATOM 1C4  CC3161    0.09 !
>>>> ATOM 1O4  OC301    -0.36 !
>>>> ATOM 2C1  CC3162    0.29 !
>>>> BOND 1O4  2C1
>>>> I have to remove the HO4, HO1 and O1 lines and modify the values for the C4, O4 and C1 atoms. Then, I need to add the bond of
>>>>
>>>> [ bond ]
>>>> …
>>>> O4  +C1
>>>>
>>>
>>> Correct.
>>>
>>>> Then, I need to apply  the bonds and angles parameters in the the top_all36_carb.rtf (below), into the merged.vsd file of the GROMACS CHARMM36.
>>>> !    I    J    K    L      R(IK)   T(IKJ)    PHI   T(JKL)   R(KL)
>>>> IC   1C3  1C4  1O4  2C1    1.5071  110.40  -86.30  121.00   1.3902  ! psi
>>>> IC   1C4  1O4  2C1  2O5    1.4560  121.00 -130.97  108.63   1.4470  ! phi
>>>> IC   2O5  1O4 *2C1  2C2    1.4470  108.63 -122.09  110.89   1.5316
>>>> IC   2O5  1O4 *2C1  2H1    1.4470  108.63  121.92  111.32   1.0837
>>>>
>>>> I have figured out how to implement R(IK), T(IKJ), T(JKL) and R(KL) values into the merged.vsd file, except for the PHI values. Where (and/or how) should I put it?
>>>> Am I on the right track?
>>>
>>> You should not do this.  The .vsd file is for defining virtual sites.  The IC lines are for the CHARMM program's internal coordinate builder, specifying some optimized geometry (one that the force field in total should produce, or come very close).  All the bonded parameters you need are already in the force field because they come from the corresponding .prm files.  Do not adjust bonded parameter files.
>>>
>>> -Justin
>>>
>>>>
>>>> Thanks again for your help.
>>>> MH
>>>>
>>>>
>>>>> On May 19, 2017, at 5:36 PM, Justin Lemkul <jalemkul at vt.edu <mailto:jalemkul at vt.edu>> wrote:
>>>>>
>>>>>
>>>>>
>>>>> On 5/19/17 9:57 AM, Mohammad Hassan Khatami wrote:
>>>>>> First, I am looking for 1->4 and 1->6 linkages. In the top_all36_carb.rtf file I found different linkages for beta-glucose, but non for alpha-glucose.
>>>>>> I am trying to make a simple chain with 1->4 linkages like below:
>>>>>>
>>>>>> alpha-D-glucose,1->4,alpha-D-glucose,1->4,alpha-D-glucose,1->4,alpha-D-glucose.
>>>>>>
>>>>>
>>>>> Linkages are not specific to the sugar; most are totally generic.  A few comments suggest specific usage and may be corner cases, but your patches will be among 14aa, 14ab, 14ba, 14bb.
>>>>>
>>>>>>
>>>>>> Then, I might need to branch them with1->6 linkage.
>>>>>
>>>>> Also totally possible.
>>>>>
>>>>>> I tried Glycan Reader, but itstill crashes.
>>>>>>
>>>>>
>>>>> Uploading a correctly named PDB file should work in Glycan Reader or the Quick MD Simulator, but "still crashes" is not diagnostic of anything.  Specific help with CHARMM-GUI should be brought to their attention, though.
>>>>>
>>>>> -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

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