[gmx-users] Partial incompatibility of pdb2gmx with 4 character residue names?

[Mark Abraham]

It's not clear whether this is or should be an intended behaviour.
There are lots of considerations of atom and residue renaming that
pdb2gmx tries to handle, and it is not too surprising that there are
corner cases that don't work as everyone expects. I couldn't even find
the code where this might happen. You might have been able to avoid
renaming your .rtp entries manually by using pdb2gmx -rtpres.

Where would have been a good place for you to have read "Using
three-letter .rtp residue names might make life easier. Consider using
pdb2gmx -rtpres if you need longer names."

Mark

On Thu, Aug 15, 2013 at 3:45 PM, Joerg Sauter <joerg.sauter at mpikg.mpg.de> wrote:
> Hi,
>
> I am building a tetramer of beta(1-4) linked glucose with pdb2gmx (Gromacs
> 4.6.3). I have three residue types BGLB BGLM BGLE defined in the
> aminoacids.rtp with which I want to built the chain BGLB BGLM BGLM BGLE.
>
> Using pdb2gmx works fine, until I link the residues via the specbond.dat
> file:
>
> 3
> BGLB    O4        1        BGLM    C1        1        0.147 BGLB    BGLM
> BGLM    O4        1        BGLM    C1        1        0.147 BGLM    BGLM
> BGLM    O4        1        BGLE    C1        1        0.147 BGLM    BGLE
>
> In this case I get the error message:
>
> Fatal error:
> Atom HO4 in residue BGLE 4 was not found in rtp entry BGLM with 21 atoms
> while sorting atoms.
>
> Using the -debug 1 mode I found that internally (in the chain.pdb) the
> residue names get shortened to BGL thus making the residues
> indistinguishable. Therefore, pdb2gmx seems to mix up residues resulting in
> the error above. If I use residue names consisting of only 3 characters
> everything works fine again.
>
> Is this behaviour known or intended? I could not find any note on this in
> the manual and it took me some time to figure this out, so maybe there
> should be at least a warning.
>
> Best,
> Joerg
>
> --
> 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/Support/Mailing_Lists/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/Support/Mailing_Lists