[gmx-users] Strange error message when trying to preprocess a cyclotide in GROMACS

[Justin Lemkul]

On 6/26/14, 4:28 PM, Matthew Stancea wrote:
> Hello again!
>
> Because of the help of Dr. Lemkul and Dr. Abraham, I was able to successfully generate a gromacs topology of a cyclotide from a pdb file!
>
> (The command I used was "pdb2gmx_mpi -ff amber99sb -water tip3p -f 1NB1.pdb -o conf.gro -p topol.top -i posre.itp -inter -ter".)
>
> In the generated file, the only issue was that there were a total of three hydrogens bonded to the beginning nitrogen and a total of two oxygens bonded to the ending carbon. Since all of the other information was correct, I did these steps in this order so that I could generate a completely correct cyclotide gromacs topology:
>
>
> First, I visualized the structure using vmd and took note of which two of the three hydrogens needed to be removed and which oxygen of the two needed to be removed.
>
> Second, I edited the conf.gro file and the topol.top file, and I removed the three lines correspondin to the extraneous atoms.
>
> Third, I used a python script to rename the conf.gro atom numbers.
>
>
> Fourth, I used another python script to simultaneously rename the atom numbers for the section "[ atoms ]", create an index for the previous number and the new number, and use that index to renumber the sections "[ bonds ]", "[ pairs ]", "[ angles ]", and "[ dihedrals ]". (I had no entries for the second section of dihedrals.)
>
>
> Lastly, after attempting to preprocess, I discovered some of the entries in "[ bonds ]" and  "[ angles ]" were renumbered twice (therefore containing erroneous connections), and so I manually renumbered them to reference the correct atoms.
>

Well, that likely means your script did something buggy, so you should be very 
wary of the rest...

>
> After this process, I once again attempted to preprocess, but this time I received the following message:
>
>
> NOTE 1 [file grompp.mdp]:
>    leapfrog does not yet support Nose-Hoover chains, nhchainlength reset to 1
>
> Setting the LD random seed to 17692
> Generated 2211 of the 2211 non-bonded parameter combinations
> Generating 1-4 interactions: fudge = 0.5
> Generated 2211 of the 2211 1-4 parameter combinations
>
> ERROR 1 [file topol.top, line 1839]:
>    No default Angle types
>

This could be a product of the aforementioned scripting problems.  Check out 
line 1839, and see what atoms those numbers map to.  Decide whether or not there 
should really be an angle between those atoms.  Likely not.

> Excluding 3 bonded neighbours molecule type 'Protein_chain_A'
> turning all bonds into constraints...
>
> NOTE 2 [file topol.top, line 3598]:
>    System has non-zero total charge: 0.454300
>    Total charge should normally be an integer. See
>    http://www.gromacs.org/Documentation/Floating_Point_Arithmetic
>    for discussion on how close it should be to an integer.
>

Here is a critical problem.  Simply deleting atoms is not a correct approach on 
its own, because you're altering the chemical nature of the molecule.  Atom 
types and charges will change because your turning a free amine into an amide, 
which may also have implications for the alpha carbon, and, as is the case in 
AMBER force fields, the entirety of the residue.  You need to completely change 
the topology entries for the first and last residues to account for this 
"simple" change.

>
>
> Warning: atom name 2 in topol.top and conf.gro does not match (H1 - H)
>
> WARNING 1 [file topol.top, line 3598]:
>    1 non-matching atom name
>    atom names from topol.top will be used
>    atom names from conf.gro will be ignored
>

Problem in mapping parameters; this means something is out of order.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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