[gmx-users] CGenFF generated prm/itp overriding and messing with charmm36-nov2016.ff?

Mon Mar 13 01:21:15 CET 2017

Sorry for the delay, I didn't see this email.

I don't believe it is a problem with gromacs, but a problem with
cgenff_charmm2gmx.py from (from
http://mackerell.umaryland.edu/download.php?filename=CHARMM_ff_params_files/cgenff_charmm2gmx.py
)

cgenff_charmm2gmx.py wrote the .pdb atom names from columns 14-17 instead
of 13-16. This code works fine if the atom names in .str outputted by
https://cgenff.paramchem.org/initguess/ is <=3 chars long. If it is 4 chars
long (one char in the atom name will be in column 17, which isn't read by
gromacs, and it shouldn't) the gromacs errs because the gmx outputed pdbs
(from editconf, solvate, the the like)  does not match the atom name in
.top.

I have contacted them about this error.

Thanks,
Jonathan

On Fri, Mar 3, 2017 at 5:02 AM, Justin Lemkul <jalemkul at vt.edu> wrote:

>
>
> On 3/2/17 11:28 PM, Jonathan Saboury wrote:
>
>> I think I narrowed it down to the problem. This time I only simulated MIY
>> in water and the problem occurs in gmx solvate and gmx grompp
>>
>> gmx solvate seems to only reads a max of 3 chars for atom type/number (not
>> sure the correct name, it is the 3rd entry in pdb). This results in the
>> output pdb invalid if the atom type/name has 3+ chars (4 in the case of
>> the
>> cgenff generated itps.
>>
>> Here is the comparison below:
>> terminal IO: http://pastebin.com/raw/Vzibp4U0
>> miy_b4_solv.pdb: http://pastebin.com/raw/CG8S3scz
>> miy_b4_ion.pdb: http://pastebin.com/raw/1StxheeJ
>>
>> Fixing the atom types does not solve this problem however (grompp still
>> only reading 3 chars?).
>> So either gmx solvate/grompp  needs to fix their atom type reading or
>> cgenff needs to fix their atom naming?
>>
>> Just wanted to confirm this is the correct assessment of the  problem with
>> you before proceeding to ask for changes :)
>>
>>
> I don't follow.  How does the spacing of atom names (note: there are not
> "types" in a coordinate file, that's a topological construct) affect the
> contents of the generated .str/.itp file?
>
> If we've got a bug in our conversion script, I'm happy to fix it.  I just
> need to know what I'm fixing.
>
> -Justin
>
>
> Thanks as always!
>>
>> - Jonathan
>>
>> On Thu, Mar 2, 2017 at 2:57 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>>
>>
>>>
>>> On 3/2/17 5:55 PM, Jonathan Saboury wrote:
>>>
>>> Dear Justin
>>>>
>>>> Do you mean the "Include parameters that are already in CGenFF" option?
>>>> I
>>>> didn't tick that option but I did have to tick "guess bond orders from
>>>> connectivity" or else it would error because of hypervalent carbon (5
>>>> bonds)
>>>>
>>>>
>>>> Yeah, that's the box.  The only way I can think of to get overriding
>>> parameters is to include things that are already present in the force
>>> field.
>>>
>>> -Justin
>>>
>>> Screen shot: http://oi65.tinypic.com/2h5tqfs.jpg
>>>
>>>>
>>>> Thanks!
>>>>
>>>> - Jonathan
>>>> <http://oi65.tinypic.com/2h5tqfs.jpg>
>>>>
>>>>
>>>>
>>>>
>>>> On Thu, Mar 2, 2017 at 2:07 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>>>>
>>>>
>>>>
>>>>> On 3/2/17 3:17 PM, Jonathan Saboury wrote:
>>>>>
>>>>> Hello all,
>>>>>
>>>>>>
>>>>>> I'm trying to run a simulation of 2DRD in water with
>>>>>> charmm36-nov2016.ff
>>>>>> (AcrB co complexed with minocycline). The main problem I'm having is
>>>>>> generating an .itp of minocycline (getting it from a reference would
>>>>>> not
>>>>>> work for me as I need to simulate other small organic molecules not in
>>>>>> literature). I have run md on 2drd not complexed successfully with the
>>>>>> same
>>>>>> methodology.
>>>>>>
>>>>>> I've tried charmm-gui and that worked but that was messy as I needed
>>>>>> to
>>>>>> include two forcefields:
>>>>>> ./charmm36-nov2016.ff/forcefield.itp and charmm36.itp. This proceeded
>>>>>> to
>>>>>> overriding Bond parameters. What parameters were overridden and the
>>>>>> effect
>>>>>> of that I have no idea (hopefully you can enlighten me).
>>>>>>
>>>>>>
>>>>>> If you are building a system with CHARMM-GUI, you should not make any
>>>>>>
>>>>> changes whatsoever to the topology it gives you.  The reason is that
>>>>> what
>>>>> CHARMM-GUI does is extract only the necessary subset of the CHARMM
>>>>> force
>>>>> field that is relevant to your system and put it in an all-included
>>>>> topology.  This is different from the "normal" GROMACS approach of
>>>>> constructing a whole force field and then using a subset of it.
>>>>>
>>>>> So I tried CGenFF. This generated the itp (along with a prm) with the
>>>>> ff
>>>>> I
>>>>>
>>>>> wanted was used, no other ff was included (yay!).
>>>>>>
>>>>>> However, this led to a "overriding Bond parameters" as well as a
>>>>>> number
>>>>>> of
>>>>>> coordinates in coordinate file (complex_b4_ion.pdb, 410866) does not
>>>>>> match
>>>>>> topology (topol.top, 434665) error. I'm fairly sure my .top file
>>>>>> includes
>>>>>> everything correctly, so I'm thinking that somehow this .prm/.itp
>>>>>> messed
>>>>>> up
>>>>>> with original ff and added some atoms. Or did I do something very
>>>>>> stupid?
>>>>>>
>>>>>> Terminal IO: http://pastebin.com/raw/KY6SRtdm
>>>>>> topol.top: http://pastebin.com/raw/fKJfpr6a
>>>>>> Whole folder download (10.9 MB): http://www.filedropper.com/2drdtar
>>>>>>
>>>>>>
>>>>>> The errors suggest you enabled the option to include all relevant
>>>>>> force
>>>>>>
>>>>> field parameters in the resulting stream file.  This will cause overlap
>>>>> with the existing force field, which is not what you want.
>>>>>
>>>>> -Justin
>>>>>
>>>>> Thank you for the help, I really appreciate it :-)
>>>>>
>>>>>
>>>>>> - Jonathan
>>>>>>
>>>>>>
>>>>>> --
>>>>>>
>>>>> ==================================================
>>>>>
>>>>> 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
>>>>>
>>>>> ==================================================
>>>>> --
>>>>> 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.
>>> 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
>>>
>>> ==================================================
>>> --
>>> 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.
> 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
>
> ==================================================
> --
> 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.
>