[gmx-developers] Generating topologies and assigning hydrogens

[Michael Shirts]

> The code analyzes hydrogen bonding (networks) in the structure to assign
> the most probable protonation state, and seems to do a proper job there.
> Indeed, it has to do with more than just topology assignment. Other than
> that, I've never even looked at this part of the code, so can't help you
> there.

Ah, that makes sense.  The problem is, if the hydrogens are already in
place (for example, if you are taking coordinates from a pdb output by
another program), it causes the die with an error, because it tries to
add in hydrogens that are already there.

Perhaps there is some hidden flag (because I don't see it documented
in pdb2gmx) that allows you to avoid this?  Or does this need to be
added so that it will accept the pdb "as is" instead of trying to
protonate?

Thanks,
Michael