[gmx-users] OXY not found

[Chris Neale]

Hi Andy,

I did some work on molecular oxygen back in 2006. There have been a few requests for molecular oxygen parameters so I decided to
dig them up. Here are the parameters that I used for that work. I leave it to you to read the relevant papers and decide if 
these parameters are appropriate for your simulations.

Note that I went with an epsilon value that I converted myself from Fisher and Lago (3.71000e-01) and considered
the value used by Marrink and Berendsen (3.63000e-01) to have been incorrectly converted. However, we never did publish our
simulations so I can't give you a reference for that discrepancy, which, to my knowledge, has not previously been reported.

After modifying the files as I outline below, you will need to rename your oxygen residue and atomic names to match and then it
should respond properly to a run through pdb2gmx.

It probably makes a lot more sense to create an otwo.itp file, but I did this in my early gromacs days and didn;t know how
to do that then. I am leaving the instructions intact as I did it at the time since that is easier for me.

For a separate set of parameters, you might look at Bembenek and Rice JCP vol 113 #6 p. 2354-9 and Bembenek JCP 124 (2006).
I did not decide that the Fisher and Lago parameters were "better", I just picked them since they had previously been used in our lab:
Al-Abdul-Wahid MS, Yu CH, Batruch I, Evanics F, Pomès R, Prosser RS. "A combined NMR and molecular dynamics study of the transmembrane 
solubility and diffusion rate profile of dioxygen in lipid bilayers."  Biochemistry. 2006 Sep 5;45(35):10719-28.
Although that paper utilized the values directly from Marrink and Berendsen.


## Add to ffX.atp:
; Added by Chris Neale March 16 2006 based on
; 1) Fisher and Lago J. Chem. Phys. 78(9), 1 May 1983
; 2) Marrink and Berendsen J. Phys. Chem. 1996, 100, 16729-16738
 OOT   15.99940  ; O in nitro R-NO2

## Add to ffX.rtp:
[ OTWO ]  
; Added by Chris Neale March 16 2005 based on
; 1) Fisher and Lago J. Chem. Phys. 78(9), 1 May 1983
; 2) Marrink and Berendsen J. Phys. Chem. 1996, 100, 16729-16738
 [ atoms ]
   OOA   OOT         0.00      0
   OOB   OOT         0.00      0
 [ bonds ]
   OOA   OOB

## Add to ffXbon.itp
; Added by Chris Neale March 16 2005 based on
; 1) Fisher and Lago J. Chem. Phys. 78(9), 1 May 1983
; 2) Marrink and Berendsen J. Phys. Chem. 1996, 100, 16729-16738
[ bondtypes ]
; i    j  func       b0          kb
 OOT   OOT      1    0.10160     659840.
;[ constrainttypes ]
;[ angletypes ]
;[ dihedraltypes ]

## Add to ffXnb.itp:
; Added by Chris Neale March 16 2006 based on
; 1) Fisher and Lago J. Chem. Phys. 78(9), 1 May 1983
; 2) Marrink and Berendsen J. Phys. Chem. 1996, 100, 16729-16738  
; name  bond_type    mass    charge   ptype          sigma      epsilon
 OOT        OOT     1   15.99940     0.000       A    3.09000e-01  3.71000e-01

Chris.


-- original message --

Hi, I run pdb2gmx (with all the ff that Gromacs bring by default) on
myoglobin (1A6M.pdb) wich I get from the "protein data bank". and I got the
following

Fatal error:
Residue 'OXY' not found in residue topology database

I check the ff and I did not found the oxigen molecule.
I saw that somebody allready use this molecule in the mailing list:

On Tue, 18 Jun 2002, Valentin Gogonea

Is there a chance to contact him?
Is this molecule in any other ff?
How can I fix this?
How can I take the OXI and SO4 molecules out of my .pdb file without making
mistakes?
Hopping to be clear.
Thanks in advance

Andy