[gmx-users] bond managment in grompp

[francesco oteri]

Dear gromacs user,
I noted a strange behavior in topology generation workflow.

I have a protein whose active site contains a Ni atom linked to 4 Cys.
I inserted the 4 bonds by specbonds.dat.   When I run pdb2gmx, I have
inspected
the Linking messages and they fit with the expected topology, I mean
pdb2gmx
correctly detects the four bonds nickel-sulphur.
So I run grompp and the topol.tpr has been correctly generated.

Then I compared topol.top and topol.tpr bonds and I noticed that in
topol.top there a lot
of fake-bonds, i.e Ni-hydrogen, Ni-Carbonium.

I dumped topol.tpr through gmxdump, and I noticed that these bonds are no
more present
and only  "right" bonds are present.

So, I am wondering why the fake bonds are inserted in topol.top and how
grompp detectes
these bonds.


Francesco
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120321/2638e317/attachment.html>