[gmx-users] Re: gmx-users Digest, Vol 52, Issue 96

[Ragnarok sdf]

>
> Ragnarok sdf wrote:
> > Hello. Ie been trying to run a DNA simulation with gromacs using amber99
> > force field. I've started with my DNA sequence complexed with the
> > protein, but then decided to separate them to know exactly what the
> > problem was. Now I am not even using the DNA sequence from my pdb file.
> > I`ve used the NAB server to create a DNA sequence which is exactly the
> > same as the one I had. I decided to do this in order to minimize the
> > problems with the amber forcefield. Well, that didn't work either.
> > I keep getting the "Error - No default Bond types.....no default angle
> > types....etc "
> > On a previous email I asked the same question, and the answer was. Take
> > a look at the lines where the error appears. Just to be more specific on
> > my problem.....ERROR 0 [file "1vkxdna_C.itp", line 71]:
> >   No default Bond types
> > ERROR 0 [file "1vkxdna_C.itp", line 196]:
> >   No default Angle types
> > Ok. I've taken the last advice and went ther to line 196. Except that on
> > line 196 there are 4 columns
> > "16    18    19     1"  with values that I cannot spot the problem. I
> > went to the manual to see what each line means but came out without
> > knowing exactly what to do with them.
>
> The listing there are the three atoms in the angle (ai, aj, ak) and the
> function
> type.  Descriptions of these are in Chapter 5 of the manual.  Details are a
> bit
> sparse regarding angles, but the text regarding other parameters and
> topology
> format are more detailed, and generally applicable.
>
> Other lines within the [ angles ] section should have angle parameter
> information following the fourth column, but that line will not.  Determine
> if
> there should be, or if something else has gone wrong (see below).
>
> > Another problem I've spotted is that the charge (no matter what i try)
> > always comes out as a non integral value. I've renamed all atoms exactly
> > like they should be...one by one....really...one by one.
> > I would like some advice on the problem if possible.
>
> Then something probably went wrong with topology creation, but that's very
> hard
> to diagnose.  Each residue (nucleotide) in the .top should end with an
> integer
> charge (;qtot in the .top).  See if this is breaking down, and if so, post
> one
> of the nucleotides on the list if you can't find out what's wrong with it.
>
> Also, please post your pdb2gmx command line from when you created the .top.
>
> -Justin
>
> >
> > Ps: The amber Dickerson test works out just fine.
> >
> >
> > ------------------------------------------------------------------------
> >
>
Tahnk you Justein for the reply.
I read again chapter 5, this time I tried to be more thorough, but still I
don't really understand why my topology is coming out bad. My .itp files do
not look like the one from the manual abou UREA. What i mean is, the angles
and bonds parts of the file have only 3 columns. Although I read the manual,
I am still not sure about the real problem here.
You mentioned that each residue should end up with an integer charge. Well,
mine don't.
What exactly do you mean with "post one of your nucleotides here"? From my
pdb file or from my itp generated files? The entire double helix with all
nucleotides or really just one of them?
Sorry for asking these details. But I just want to make sure I don't waste
your time with useless posts.
Thank you again
And thank you in advance
Fabrício
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20080823/2671431d/attachment.html>