VMD-L Mailing List

From: João Ribeiro (jribeiro_at_ks.uiuc.edu)
Date: Mon Nov 12 2018 - 09:35:10 CST

Hi Francesco,

Sorry for the delay in responding to your email. If the molecule that
you are simulating was simulated before, I would suggest find the
publication and see how the authors prepared the system and if new
parameters were published. You might need to parameterize your
molecule (entirely or partially), and if someone already proposes
something similar, you can (and should) evaluate if the molecule is
close enough to yours.

I hope this helps

Best

João

On Tue, Oct 23, 2018 at 8:51 AM Francesco Pietra <chiendarret_at_gmail.com> wrote:
>
> Hi Joao:
> As you suspected, I forgot to change the topology (C8 CG2D1), although the parameter .str file was correct (C8 CG321).
>
> Amended, however a problem arose on minimization
> UNABLE TO FIND BOND PARAMETERS FOR CG314 NR3 (ATOMS 6715 3431)
> 6715 is C9 of the ligand
> 3431 is NE2 of HSP
>
> While nothing was found with CGenFF, with charmm27 someone has proposed for methylated histidine (https://www.charmm.org/ubbthreads/ubbthreads.php?ubb=showthreaded&Number=35082)
>>
>> BOND
>> NR3-CT3 261.0 1.49
>> CT3-HA3 322 1.110
>> ANGLE
>> NR3-CT3-HA3 51.5 107.5
>
> DIHE
> NR3-CPH2-NR3-CT3 0 2 180
> HR2-CPH2-NR3-CT3 0 2 180
> CPH2-NR3-CT3-HA3 0 0 0
> HR1-CPH1-NR3-CT3 0 2 180
> CPH1-NR3-CT3-HA3 0 0 0
> CPH1-CPH1-NR3-CT3 0 0 0
>
>
> Any better idea? Or calculation of the force constant can not be avoided?
>
> thanks
> francesco
>
>
> On Mon, Oct 22, 2018 at 7:53 PM João Ribeiro <jribeiro_at_ks.uiuc.edu> wrote:
>>
>> Hi Francesco,
>>
>> Unless you changed your topology file, the C8 was defined (in your first email) as "ATOM C8 CG2D1 0.098". I don't know how your corrected patch looks like but you can change the atom type in the patch, like "ATOM 1C8 CG321 -0.124 "
>>
>> Whenever you have to change the atom type of a structure, please do it at the psfgen level, not changing the psf file.
>>
>> Best
>>
>> João
>>
>>
>> On Mon, Oct 22, 2018 at 12:08 PM Francesco Pietra <chiendarret_at_gmail.com> wrote:
>>>
>>> Sorry, (I tried correcting the .psf file by replacing CG2D1 with CG32), should be read (I tried correcting the .psf file by replacing CG2D1 with CG321).
>>> f
>>>
>>> ---------- Forwarded message ---------
>>> From: Francesco Pietra <chiendarret_at_gmail.com>
>>> Date: Mon, Oct 22, 2018 at 7:03 PM
>>> Subject: Re: vmd-l: psfgen patching protein-ligand
>>> To: <jribeiro_at_ks.uiuc.edu>
>>> Cc: VMD Mailing List <vmd-l_at_ks.uiuc.edu>
>>>
>>>
>>> Hi
>>> Previously I answered OK, while now problems emerge with minimization.
>>> psfgen assigned atom types to the ligand
>>> C7 (CG205 carbonyl C)
>>> C8 (CG2D1 alkene; RHC=; imine C))
>>> C9 (CG314 aliphatic C with 1H, adjacent to N+)
>>>
>>> O H H
>>> || | |
>>> C7-C8-C9-C10
>>> |
>>> H
>>>
>>> C9 assignment is correct, as this is for bond with HSP NE2, following the patching indications.
>>> C7 assignment is also correct.
>>>
>>> C8 assignment is making wrong chemistry, it should be CG321, while the .str param file for the ligand reads accordingly
>>> ATOM C8 CG321 -0.124 ! 2.500
>>>
>>> Therefore, minimization crashes with error: no bond parameters for CG2D1 CG205
>>>
>>> What to do? I can not assume any bug in psfgen. I tried correcting the .psf file by replacing CG2D1 with CG32, which raised other major problems.
>>>
>>> Thanks for advice
>>>
>>> francesco
>>>
>>> On Tue, Oct 16, 2018 at 5:56 PM João Ribeiro <jribeiro_at_ks.uiuc.edu> wrote:
>>>>
>>>> Hi,
>>>>
>>>> Sometimes is useful to check other patches in the CHARMM distribution. For instances, the DISU patch below, you need to list from which residue you are deleting the atoms, and from which residues belong the atoms that you are binding. The syntax is the residue number (1 or 2) before the atom's name and it matches the order of the residues that you provided in the command.
>>>>
>>>> PRES DISU -0.36 ! patch for disulfides. Patch must be 1-CYS and 2-CYS.
>>>> ! use in a patch statement
>>>> ! follow with AUTOgenerate ANGLes DIHEdrals command
>>>>
>>>> GROUP
>>>> ATOM 1CB CT2 -0.10 !
>>>> ATOM 1SG SM -0.08 ! 2SG--2CB--
>>>> GROUP ! /
>>>> ATOM 2SG SM -0.08 ! -1CB--1SG
>>>> ATOM 2CB CT2 -0.10 !
>>>> DELETE ATOM 1HG1
>>>> DELETE ATOM 2HG1
>>>> BOND 1SG 2SG
>>>>
>>>> I hope this helps
>>>>
>>>> Best
>>>>
>>>> João
>>>>
>>>>
>>>>
>>>> On Tue, Oct 16, 2018 at 10:47 AM Francesco Pietra <chiendarret_at_gmail.com> wrote:
>>>>>
>>>>> Hi Joao
>>>>> Thanks a lot for correcting my PRES. Now no warning in the log file about either patching or improper setting, however, neither atom H91 at the methylene carbon of the ligand, nor HE2 of HSP were removed, and the intended bond was not formed.
>>>>>
>>>>> Correction to my PRES reads
>>>>> ..........
>>>>> DELETE ATOM H91
>>>>> DELETE ATOM HE2
>>>>> BOND C9 NE2
>>>>> ! IMPR NR3 CPH1 CPH2 CG314
>>>>> ! IMPR NR3 CPH2 CPH1 CG314
>>>>> IMPR NE2 CD2 CE1 C9
>>>>> IMPR NE2 CE1 CD2 C9
>>>>>
>>>>> Since a long time I had no need of patching. It seems that I forgot the rules.
>>>>> Thanks for advice
>>>>>
>>>>> francesco
>>>>>
>>>>> On Tue, Oct 16, 2018 at 5:21 PM João Ribeiro <jribeiro_at_ks.uiuc.edu> wrote:
>>>>>>
>>>>>> Hi Francesco,
>>>>>>
>>>>>> The definition of your improper angles should be done using atoms' name (e.g NE2) and not type (e.g. NE3).
>>>>>>
>>>>>> Best
>>>>>>
>>>>>> João
>>>>>>
>>>>>> On Tue, Oct 16, 2018 at 10:13 AM Francesco Pietra <chiendarret_at_gmail.com> wrote:
>>>>>>>
>>>>>>> Hi
>>>>>>> I can't understand the error message in trying to bind the sp3 carbon C9 of a ligand to the NE2 atom of HSP residue, which implies deleting the HE2 atom of HSP and the H91 atom at C9 of the ligand.
>>>>>>>
>>>>>>>> PRES ZAHI 0.661 ! patch for ligand to HSP bonding
>>>>>>>> ! use in patch statement
>>>>>>>> ! follow with AUTOgenerate ANGles DIHEdrals command
>>>>>>>> GROUP !
>>>>>>>> ATOM C10 CG311 0.303 !
>>>>>>>> ATOM C9 CG314 0.17 !
>>>>>>>> ATOM H91 HGA1 0.09 !
>>>>>>>> ATOM C8 CG2D1 0.098 ! CD2--NE2--CE1
>>>>>>>> GROUP ! /
>>>>>>>> ATOM CD2 CPH1 0.19 ! C10--C9--C8
>>>>>>>> ATOM NE2 NR3 -0.51 ! |
>>>>>>>> ATOM CE1 CPH2 0.32 ! H91
>>>>>>>> DELETE ATOM H91
>>>>>>>> DELETE ATOM HE2
>>>>>>>> BOND C9 NE2
>>>>>>>> IMPR NR3 CPH1 CPH2 CG314
>>>>>>>> IMPR NR3 CPH2 CPH1 CG314
>>>>>>>
>>>>>>>
>>>>>>> the error message reads
>>>>>>>
>>>>>>> psfgen) no atom NR3 in residue HSP:227 of segment PROD
>>>>>>>
>>>>>>> and therefore patching failed, while the .psf file shows that NE2 of HSP is a NR3 type atom.
>>>>>>>
>>>>>>> My PRES should be wrongly set. Thanks for advice.
>>>>>>>
>>>>>>> francesco pietra
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> ……………………………………………………...
>>>>>> João Vieira Ribeiro
>>>>>> Theoretical and Computational Biophysics Group
>>>>>> Beckman Institute, University of Illinois
>>>>>> http://www.ks.uiuc.edu/~jribeiro/
>>>>>> jribeiro_at_ks.uiuc.edu
>>>>>> +1 (217) 3005851
>>>>
>>>>
>>>>
>>>> --
>>>> ……………………………………………………...
>>>> João Vieira Ribeiro
>>>> Theoretical and Computational Biophysics Group
>>>> Beckman Institute, University of Illinois
>>>> http://www.ks.uiuc.edu/~jribeiro/
>>>> jribeiro_at_ks.uiuc.edu
>>>> +1 (217) 3005851
>>
>>
>>
>> --
>> ……………………………………………………...
>> João Vieira Ribeiro
>> Theoretical and Computational Biophysics Group
>> Beckman Institute, University of Illinois
>> http://www.ks.uiuc.edu/~jribeiro/
>> jribeiro_at_ks.uiuc.edu
>> +1 (217) 3005851 

-- 
……………………………………………………...
João Vieira Ribeiro
Theoretical and Computational Biophysics Group
Beckman Institute, University of Illinois
http://www.ks.uiuc.edu/~jribeiro/
jribeiro_at_ks.uiuc.edu
+1 (217) 3005851