[gmx-users] H2 topology

Thu Feb 4 14:20:06 CET 2010

On 04/02/10 21:01, 011013021-Jyotsna wrote:
> Dear Mark,
> Your suggestions were very helpful .
> As per the literature I am following , It is required that the dummy
> atom has mass 0 and charge -0.950.Hence , I created the following itp
> file with this in mind.
> ######
> ; topology for hydrogen.
>
> [ moleculetype ]
> ;name nrexcl
> H2 2
>
> [ atoms ]
> ; nr type resnr residu atom cgnr charge mass ; total charge
> 1 H 1 H2 H1 1 0.475 1.00800 ; 0.00
> 2 H 1 H2 H2 1 0.475 1.00800 ; 0.00
> 3 DUM 1 H2 DUM 1 -0.950 0.00000 ; 0.00
>
> [ virtual_sites2 ]
> ; Site from funct a
> 3 1 2 1 0.074

That gets you an asymmetric dummy atom, which seems weird. David told 
you how to get a symmetric one. The manual also makes that clear.

>
> [ constraints ]
> 1 2 2 0.074
> ######
>
>
> But during energy minimization I encountered many problems.

Strictly EM is done by mdrun. Your inputs to grompp were invalid, so EM 
never commences.

> The errors are:
> ########### error ##########
> Ignoring obsolete mdp entry 'cpp'
>
> Back Off! I just backed up em_ion_out.mdp to ./#em_ion_out.mdp.9#
> checking input for internal consistency...
> processing topology...
> Opening library file /usr/local/gromacs/share/gromacs/top/ff_dum.itp
> Generated 279 of the 1225 non-bonded parameter combinations
> Opening library file /usr/local/gromacs/share/gromacs/top/spc.itp
> Opening library file /usr/local/gromacs/share/gromacs/top/ions.itp
> Excluding 3 bonded neighbours molecule type 'Protein_A'
> Excluding 3 bonded neighbours molecule type 'Protein_B'
> Excluding 2 bonded neighbours molecule type 'SOL'
> Excluding 2 bonded neighbours molecule type 'H2'
> processing coordinates...
> double-checking input for internal consistency...
> Cleaning up constraints and constant bonded interactions with virtual sites
> renumbering atomtypes...
> converting bonded parameters...
>
> ERROR 1 [file 2frv_ion.top, line 40]:
> atom DUM (Res H2-6) has mass 0
>
>
>
> ERROR 2 [file 2frv_ion.top, line 40]:
> atom DUM (Res H2-7) has mass 0
>
>
>
> -------------------------------------------------------
> Program grompp_d, VERSION 4.0
> Source code file: grompp.c, line: 940
>
> Fatal error:
> There were 2 errors in input file(s)
> ----------------------------------------------
> ####### error ########
> Is this due to a problem in the .itp file?

Yep. Dummy atoms need mass, or what is the point?

Mark

> Even though i referred to the previous queries made in the forum , I was
> not able to get an answer,
> Please guide me as to where I have gone wrong .
> Thank You,
> jyotsna
>
>
>
> On Wed, 03 Feb 2010 22:42:09 +1100
> Mark Abraham <mark.abraham at anu.edu.au> wrote:
>> *This message was transferred with a trial version of CommuniGate(r) Pro*
>>
>>
>> ----- Original Message -----
>> From: 011013021-Jyotsna <011013021 at bioinfo.sastra.edu>
>> Date: Wednesday, February 3, 2010 21:48
>> Subject: Re: [gmx-users] H2 topology
>> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>>
>>>
>>> Dear David,
>>>
>>> Thanks for your reply. From your reply, I get the impression that the
>>> bond length of my hydrogen molecule gets adjusted duing minimization
>>> of the whole system ( protien+water+ inserted hydrogen molecule). But
>>> the issue impending is the insertion of the hydrogen molecule itself
>>> into the protein+water box created in previous step. For
>>> insertion/addition of hydrogen molecules I used the command:
>>>
>>> genbox_d -cp 2frv.conf.gro -o 2frv.solv.gro -p 2frv.top –ci h2.gro
>>> –nmol 100
>>
>> What about a two-stage process? Do genbox -ci h2.gro -nmol 100 when
>> there's no water, and then solvate *afterwards*.
>>
>> Or, generate solvent, and get a list of 100 random numbers in the
>> right range, manually delete those water molecules to create
>> interstices, update your [molecules] section, then use genconf -ci and
>> equilibrate carefully to fix the density.
>>
>> Or, generate solvent for a box somewhat smaller than the one you want,
>> use editconf -scale to scale the coordinates up to create interstices,
>> use genconf -ci and then minimize really gently to fix all the bond
>> lengths, and equilibrate carefully to fix the density.
>>
>> Mark
>>
>>> It reports to me the inability to add any of the hydrogen molecules,
>>> ie it adds 0 molecules out of 100 requested.
>>> ###########error########################
>>> Reading molecule configuration
>>> Gallium Rubidium Oxygen Manganese Argon Carbon Silicon
>>> Containing 3 atoms in 1 residue
>>> Initialising van der waals distances...
>>> Try 999
>>> Added 0 molecules (out of 100 requested) of H2
>>> Writing generated configuration to 2frv.solv.gro
>>>
>>> Back Off! I just backed up 2frv.solv.gro to ./#2frv.solv.gro.1#
>>> PERIPLASMIC HYDROGENASE; PERIPLASMIC HYDROGENASE
>>>
>>> Output configuration contains 8340 atoms in 977 residues
>>> Volume : 1299.08 (nm^3)
>>> Density : 143.043 (g/l)
>>> Number of SOL molecules: 186
>>>
>>> Processing topology
>>> Removing line #40 'SOL 39308' from topology file (2frv.top)
>>> ####################error end###############
>>> How do i rectify this problem?
>>> I suspect the problem is due to the bond length of the hydrogen
>>> molecule that I created with 1.66A distance ( as described in my
>>> previous post) and the fact that the gap between the solvent
>>> molecules (in my case , water) is not big enough to accomodate the H2
>>> molecules(as I had used the pdb of CO2 and replaced each O with an H
>>> and C atom with dummy atom.). So could you please help me figure out
>>> any tool that helps in adjusting the bond length of my hydrogen
>>> molecule in pdb to the desired length of 0.7A.
>>>
>>> Or as you said is it possible to minize a single molecule using
>>> md_run program, as I encountered certain problems in doing so .
>>>
>>> Sorry for the lengthy mail.
>>>
>>> Thank you ,
>>> Jyotsna
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Wed, 03 Feb 2010 09:30:21 +0100
>>> David van der Spoel <spoel at xray.bmc.uu.se> wrote:
>>> >*This message was transferred with a trial version of CommuniGate(r)
>>> Pro*
>>> >On 2/3/10 7:52 AM, 011013021-Jyotsna wrote:
>>> >>
>>> >>Dear Justin ,
>>> >>
>>> >>Thank you very much for your help.
>>> >>To create a linear H2 molecule with a dummy atom in the center is
>>> >>proving to be quite a difficult task.
>>> >>Hence , I created a pdb file of CO2 using Cerius(2) , and replaced the
>>> >>atoms in the pdb file with 2 H atoms and a dummy atom.(since CO2 is a
>>> >>three atom , linear system i.e the two O atoms replaced by two H atoms
>>> >>and the C atom by the dummy)
>>> >>The problem is the bond distance. For CO2 , it is 1.66 A but I need
>>> the
>>> >>distance for H2 to be only 0.7 A as per the literature I am
>>> following.>>By replacing the atoms in the pdb file , there is a
>>> problem regarding
>>> >>distance since the coordinates of the pdb also point towards 1.6 A.
>>> >>I would be very grateful if you could tell me how I can build a PDB
>>> for
>>> >>H2 with dummy atom with the right distance.
>>> >>
>>> >>Thank You,
>>> >>Jyotsna
>>> >Doesn't matter if the pdb is wrong initially, you just do an energy
>>> minimization. What is important is (put in exact bond length):
>>> >
>>> >[ bonds ]
>>> >1 2 1 0.74 500000
>>> >
>>> >[ vsites2 ]
>>> >3 1 2 1 0.5
>>> >
>>> >
>>> >
>>> >-- >David van der Spoel, Ph.D., Professor of Biology
>>> >Dept. of Cell & Molec. Biol., Uppsala University.
>>> >Box 596, 75124 Uppsala, Sweden. Phone: +46184714205. Fax:
>>> +4618511755.>spoel at xray.bmc.uu.se spoel at gromacs.org
>>> http://folding.bmc.uu.se>-- >gmx-users mailing list
>>> gmx-users at gromacs.org
>>> >http://lists.gromacs.org/mailman/listinfo/gmx-users
>>> >Please search the archive at http://www.gromacs.org/search before
>>> posting!
>>> >Please don't post (un)subscribe requests to the list. Use the www
>>> interface or send it to gmx-users-request at gromacs.org.
>>> >Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>>
>>> --
>>> gmx-users mailing list gmx-users at gromacs.org
>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>> Please search the archive at http://www.gromacs.org/search before
>>> posting!
>>> Please don't post (un)subscribe requests to the list. Use the www
>>> interface or send it to gmx-users-request at gromacs.org.
>>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>> --
>> gmx-users mailing list gmx-users at gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> Please search the archive at http://www.gromacs.org/search before
>> posting!
>> Please don't post (un)subscribe requests to the list. Use the www
>> interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>