[Chimera-users] Listing hydrogen bond geometries

[Elaine Meng]

Hi Greg,
For FindHBond, Chimera "imagines" the possible locations of hydrogens given the atom types (element, hybridization state) of the heavy atoms.
<http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/findhbond/findhbond.html>

"When there are no explicit hydrogens on a potential donor atom, FindHBond will use the donor atom type to infer their presence. For functional groups where the hydrogen positions are well-determined by the positions of the nonhydrogen atoms, the H-bonds detected in the presence and absence of explicit hydrogen atoms should be virtually identical. For functional groups where the hydrogen position is not well-determined (such as rotatable hydroxyls), the presence of an explicit hydrogen atom will limit the detected H-bonds to those appropriate for the observed position, instead of all of its possible positions."

If you want explicit hydrogens, you could use Chimera's AddH tool (under Tools... Structure Editing) or addh command.  By default, it will look at the surroundings and try to place hydrogens to make hydrogen bonds.  However, it does not exhaustively optimize the H-bonding network of the whole structure.  The goal is to determine reasonable positions based on immediate surroundings  in a fairly short time.  However, if you add explicit hydrogens and then use FindHBond, there will be fewer possibilities listed, as explained above.

<http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/addh/addh.html>

For the next level of computational intensity in adding hydrogens, we recommend the "Reduce" program from the Richardson lab.   From the bottom of the AddH page,

"When a more intensive approach is desired, the program Reduce (developed by the Richardson Laboratory) is a good alternative. Reduce places hydrogens to optimize local H-bonding networks and avoid steric overlaps, while flipping certain sidechains 180 degrees as deemed appropriate to fulfill these criteria. Asparagine and glutamine sidechains may be flipped to switch their terminal N and O atoms, and the imidazole ring of histidine may be flipped to switch N and C identities. The protonation state of histidine is adjusted based on the local environment.

Reduce is available free at http://kinemage.biochem.duke.edu:

	• for on-line use as part of the MolProbity service, on a file uploaded or chosen by PDB code (individual sidechain flips can be accepted or rejected)
	• for download (from the Software section) to run on most platforms

and is described in:
Asparagine and glutamine: using hydrogen atom contacts in the choice of side-chain amide orientation. Word JM, Lovell SC, Richardson JS, Richardson DC. J Mol Biol. 1999 Jan 29;285(4):1735-47. "

I hope this helps,
Elaine

On Jul 18, 2011, at 2:16 PM, Greg Friedland wrote:

> Hi Elain,
> Thanks very much for you response and sorry for the delay in getting back to you.
> 
> I am trying to determine the potential hydrogen bonding potential (i.e. the ability to form these bonds) from hydrogen-less structures (they are from the PDB). I have been trying to model in the hydrogens but it is not trivial since there are many hydroxyls, which can rotate about their dihedrals. (if you have suggestions on how to model these hydrogens it'd be great to know about them).
> 
> So until I can get a good model with the hydrogens attached I am wondering how Chimera is able to determine potential hbonds from heavy atom positions only. Any insight you can provide would be appreciated.
> 
> Cheers,
> Greg