[gmx-users] Re: gmx-users Digest, Vol 52, Issue 109

[Justin A. Lemkul]

Ragnarok sdf wrote:
> Hi Justin, well, actually the problem is not only with grompp, since
> my non integer charges come all the way since pdb2gmx
> I used as I told:
> pdb2gmx -f 1vkxdna.pdb -o 1vkxdna.gro -p 1vkxdna.top
> editconf -f 1vkxdna.gro -o 1vkxdna_caixa.gro -c -d 0.9 -princ
> genbox -cs -cp 1vkxdna_caixa.gro -o 1vkxdna_solv.gro -p 1vkxdna.top
> grompp -f em_steep.mdp -c 1vkxdna_solv.gro -p 1vkxdna.top -o 1vkxdna_steep.tpr
> and that´s when the process stops.
> I mean, the process stops but the non integer charges were seen right
> after pdb2gmx. So probably it´s something to do with my coordinates
> file, that´s giving bad topologies.

The problem comes from aminoacids.dat, not the coordinates.  If the nucleic 
acids are present in this file, the Gromacs treats them as terminal residues and 
applies the wrong charges, according to the explanation at the ffamber site.

> Answering to your other question, about aminoacids.dat. Well I´ve
> replaced gromacs´ original file with that given by amber
> aminoacids-NA.dat. Renamed it and replaced it making a copy of the old
> one with a different name.

That's the opposite of what you want when dealing with pdb2gmx.  For other tools 
(such as analysis), you will want Gromacs to know that things like DT5, DC, etc. 
are known residues.  For the purpose of pdb2gmx, you want to hide such information.

> I understand what you meant about the treatment they suggest in the
> amber website. But how come this error appears with my DNA without
> this extra residues I´ve added and then all of the sudden they simply
> dissappear and everything works out just fine?

Probably because the bad fractional charges coincidentally add up to an integer 
when you do that.

> Plus, I´ve tried to do what they tell you to in those steps at Amber
> website, things came out worse than they were. Probably I screwed up
> something real bad, but anyway. It seems I do not have enough
> knowledge to handle this kind of problem on my own.
> Of course you can try and simulate the pdb yourself. If it is not too
> much of a problem for you.
> The pdb entrance is 1VKX. I was able to simulate the protein, but not
> the DNA fragment that comes with it. That is why I decided to create a
> DNA sequence exactly like that one and try to run sepparatelly. And
> the rest of the story you know already. Things went bad with Thymine
> and Adenine endings but things worked out fine with Cytidine and
> Guanine endings....I still don´t know why.

Using the original aminoacids.dat (the one with *no* nucleic acids), I was able 
to produce topologies for all chains of the .pdb file, protein and nucleic acid. 
  All topologies add up to integer charges.  No modifications were made to the 
.pdb file, aside from a few adjustments to residue nomenclature to coincide with 
Amber naming conventions.

> I´ll try to follow the instructions given by the AMBER website one
> more time, see if I can make things work. But first I´ll try your
> suggestion.
> You mean to replace the aminoacids.dat I am using right now (which is
> the aminoacids-NA.dat given by amber) by the old default that comes
> with gromacs?

Use the original aminoacids.dat from the standard Gromacs distribution when 
using pdb2gmx (*without* nucleic acids).

Later on, you may have to add the nucleic acids into aminoacids.dat (i.e., 
aminoacids-NA.dat) to use the other tools (probably at least starting with grompp).

Hopefully all of that makes sense.

-Justin

> Thank you
> Fabrício Bracht
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.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
> 

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

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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