[gmx-users] Getting ligand's topology

Sun Aug 19 21:58:56 CEST 2018

On 8/19/18 10:20 AM, RAHUL SURESH wrote:
> Hi.
>
> First, I feel grimaces users may not entertain other discussions in grimace
> forum. You can directly mail me if it’s something apart from gromacs.
>
> Then, I am not sure about other servers.
>
> Looking at your str file, I would say that it definitely need some work on
> our molecule. As I said before try zinc database.

I don't see how the ZINC database is relevant. Either the .mol2 
submitted to the CGenFF server is valid or it is not.

CGenFF is nice in that it tells you the potential problems with the 
ligand topology. AFAIK, no other servers do. You get a "black box" 
output that you're supposed to trust. The areas pinpointed by CGenFF 
should be examined carefully. The parameters may actually be fine, but 
the penalties are there to tell the user when there is a functional 
group that is not well described by existing molecules in the CGenFF 
database, from which the analogies are made.

There is a CGenFF tutorial available online that walks a user through 
the whole process. It requires subdividing a molecule into units that 
can be parametrized easily; large molecules should always be broken down 
into manageable pieces that have all the necessary degrees of freedom.

The tutorial materials can be accessed here: 
http://mackerell.umaryland.edu/~kenno/cgenff/download.php

CHARMM/CGenFF parametrization assumes some familiarity with QM 
calculations (geometry optimizations, potential energy scans, 
interaction energies) but the methodology is published in great detail, 
and the CGenFF paper itself is a worked example of how to parametrize a 
molecule. While the developers have made every effort to make the 
methodology publicly available, ligand parametrization and refinement is 
still an advanced concept that is best suited for experienced users, as 
knowledge of fundamental principles of MD, QM, and empirical energy 
functions is required.

-Justin

> Thank you
>
>
> On Sun, 19 Aug 2018 at 1:37 PM, Mahdi Sobati Nezhad <
> mahdisobatinezhad at gmail.com> wrote:
>
>> thanks.
>> if I use MATCH server or swissparam, I can trust to their results?!
>>
>> And this is my error when I use CGenFF:
>> "readmol2 warning: non-unique atoms were renamed. Now processing molecule
>> mae ..."
>>
>> And this is my output of penalty:
>>
>> * Toppar stream file generated by * CHARMM General Force Field (CGenFF)
>> program version 2.2.0 * For use with CGenFF version 4.0 * read rtf card
>> append * Topologies generated by * CHARMM General Force Field (CGenFF)
>> program version 2.2.0 * 36 1 ! "penalty" is the highest penalty score of
>> the associated parameters. ! Penalties lower than 10 indicate the analogy
>> is fair; penalties between 10 ! and 50 mean some basic validation is
>> recommended; penalties higher than ! 50 indicate poor analogy and mandate
>> extensive validation/optimization. RESI mae 0.000 ! param penalty= 198.400
>> ; charge penalty= 142.287 GROUP ! CHARGE CH_PENALTY ATOM C1 CG2R61 0.227 !
>> 0.000 ATOM C2 CG2R61 -0.117 ! 2.688 ATOM C3 CG2R61 -0.003 ! 5.191 ATOM C4
>> CG2R61 -0.117 ! 2.688 ATOM C5 CG2R61 0.227 ! 0.000 ATOM C6 CG2R61 0.217 !
>> 0.000 ATOM H1 HGR61 0.115 ! 0.000 ATOM H2 HGR61 0.115 ! 0.000 ATOM O1 OG301
>> -0.391 ! 0.000 ATOM O2 OG301 -0.391 ! 0.000 ATOM O3 OG301 -0.391 ! 0.000
>> ATOM C7 CG331 -0.100 ! 0.000 ATOM H3 HGA3 0.090 ! 0.000 ATOM H4 HGA3 0.090
>> ! 0.000 ATOM H5 HGA3 0.090 ! 0.000 ATOM C8 CG331 -0.100 ! 0.000 ATOM H6
>> HGA3 0.090 ! 0.000 ATOM H7 HGA3 0.090 ! 0.000 ATOM H8 HGA3 0.090 ! 0.000
>> ATOM C9 CG331 -0.100 ! 0.000 ATOM H9 HGA3 0.090 ! 0.000 ATOM H10 HGA3 0.090
>> ! 0.000 ATOM H11 HGA3 0.090 ! 0.000 ATOM C10 CG3C51 0.317 ! 142.287 ATOM N1
>> NG3C51 -0.343 ! 101.066 ATOM C11 CG321 -0.072 ! 70.919 ATOM H12 HGA2 0.090
>> ! 2.500 ATOM H13 HGA2 0.090 ! 2.500 ATOM C12 CG2R53 0.464 ! 90.248 ATOM N2
>> NG2R50 -0.515 ! 21.469 ATOM N3 NG3C51 -0.417 ! 87.522 ATOM H14 HGP1 0.341 !
>> 3.424 ATOM C13 CG321 0.028 ! 26.713 ATOM H15 HGA2 0.090 ! 2.659 ATOM N4
>> NG3N1 -0.689 ! 58.305 ATOM C14 CG2R61 0.140 ! 19.097 ATOM C15 CG2R61 -0.110
>> ! 13.488 ATOM C16 CG2R61 0.036 ! 13.718 ATOM C17 CG2R61 -0.112 ! 0.000 ATOM
>> H16 HGR61 0.115 ! 0.000 ATOM C18 CG2R61 0.071 ! 0.000 ATOM C19 CG2R61
>> -0.089 ! 0.000 ATOM H17 HGR61 0.115 ! 0.000 ATOM S SG311 -0.210 ! 105.411
>> ATOM H18 HGP3 0.160 ! 4.359 ATOM H19 HGR62 0.144 ! 0.000 ATOM Cl1 CLGR1
>> -0.152 ! 2.500 ATOM Cl2 CLGR1 -0.167 ! 0.000 ATOM HXT1 HGA1 0.090 ! 5.954
>> ATOM HXT2 HGA2 0.090 ! 2.659 ATOM HN HGP1 0.394 ! 9.368 ATOM LP1 LPH 0.050
>> ! on Cl1 ATOM LP2 LPH 0.050 ! on Cl2
>>
>>
>> Thanks for your taking time
>>
>>
>> On Sat, 18 Aug 2018 22:31 RAHUL SURESH, <drrahulsuresh at gmail.com> wrote:
>>
>>> I cannot assist you much with the description of your ligand. But what I
>>> have seen is, ring with N shows maximum penalities in most cases.
>>>
>>> When you optimise with mp2, the structure shows no much changes.
>>> So penalty is same most time.
>>>
>>> Try to download file from Zinc15 database which is most preferred file
>>>   cgenff. Still the penalty may exist in cse of N in ring. It could be
>>> optimised using ffTK. Again I find your structure is pretty big which
>> will
>>> be time consuming. Or you have to adapt divide and conquer method, which
>> is
>>> going to be complicated. So it’s your wise choice always. Maybe others,
>>> especially Dr. Justin or Dr. Mark could give you a better soln.
>>>
>>> Thank you
>>>
>>> On Sat, 18 Aug 2018 at 10:17 PM, Mahdi Sobati Nezhad <
>>> mahdisobatinezhad at gmail.com> wrote:
>>>
>>>> Thanks for your help.
>>>> my ligand is three rings that connects with a carbon and this rings
>> have
>>>> Nitrogen, oxygen, sulfur, carbon and Chlor...
>>>> And high penaltys are just for that ring that contains Nitrogen, carbon
>>> and
>>>> sulfur
>>>> And so on my choice is just using ffTK?!
>>>>
>>>>
>>>> On Sat, 18 Aug 2018 21:05 RAHUL SURESH, <drrahulsuresh at gmail.com>
>> wrote:
>>>>> Hi.
>>>>>
>>>>> To resolve the penalties are bit complicated. Tools like ffTK can
>> help
>>>> you
>>>>> do it. But output from ffTK is as in charmm and henceforth it take
>>>>> additional work ( was difficult for me ) to modify it to Gromacs
>>>> acceptable
>>>>> format.
>>>>>
>>>>> I personally don’t appreciate this method as it is very time
>> consuming
>>> if
>>>>> your structure is big and having penalties in dihedrals which again
>>> make
>>>> it
>>>>> complicated.
>>>>> Thank you
>>>>>
>>>>> On Sat, 18 Aug 2018 at 7:02 PM, Mahdi Sobati Nezhad <
>>>>> mahdisobatinezhad at gmail.com> wrote:
>>>>>
>>>>>> Hi
>>>>>> I'm a begginer in taking ligand's topology from CGENFF and my
>> '.str'
>>>> file
>>>>>> have some high penalty numbers. What I can do?!
>>>>>> And how I can do validation and optimization?!
>>>>>> I read the FAQ in its server but I don't understand!!!
>>>>>> Thanks
>>>>>> --
>>>>>> Gromacs Users mailing list
>>>>>>
>>>>>> * Please search the archive at
>>>>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>>>>> posting!
>>>>>>
>>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>>
>>>>>> * For (un)subscribe requests visit
>>>>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>>> or
>>>>>> send a mail to gmx-users-request at gromacs.org.
>>>>>>
>>>>> --
>>>>> *Regards,*
>>>>> *Rahul *
>>>>> --
>>>>> Gromacs Users mailing list
>>>>>
>>>>> * Please search the archive at
>>>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>>>> posting!
>>>>>
>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>
>>>>> * For (un)subscribe requests visit
>>>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>> or
>>>>> send a mail to gmx-users-request at gromacs.org.
>>>> --
>>>> Gromacs Users mailing list
>>>>
>>>> * Please search the archive at
>>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>>> posting!
>>>>
>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>
>>>> * For (un)subscribe requests visit
>>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>>> send a mail to gmx-users-request at gromacs.org.
>>> --
>>> *Regards,*
>>> *Rahul *
>>> --
>>> Gromacs Users mailing list
>>>
>>> * Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>> posting!
>>>
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>> send a mail to gmx-users-request at gromacs.org.
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

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