[Chimera-users] addH utility and PDB files

[Elaine Meng]

Hi Nihshanka,
I'll try to answer your questions in a brief manner, but see the AddH  
man page for a full description:
http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/addh/addh.html

AddH does not use a force field.  It uses the atom types and rules to  
place hydrogens.  For example, if there is an sp3-hybridized carbon,  
it knows there should be four bonds from that carbon in tetrahedral  
arrangement.  If two things are already bonded to that carbon, two  
hydrogens will be added to complete the known valence.  For some  
atoms, this completely determines the position of the hydrogens.  For  
other atoms, such as a hydroxyl group, there is a degree of freedom,  
and additional information is used to place the hydrogen.  The  
choices are "steric only" and "also consider H-bonds" - the first  
places the H to avoid clashes and the second places the H to both  
avoid clashes and form H-bonds where possible.

AddH does not put charges on the structure.  You could do that with  
Add Charge (or Dock Prep, which calls Add Charge).  HOWEVER, DelPhi  
completely ignores the radii and charges that Chimera knows about.   
DelPhi uses only the radii and charges in the parameter files that  
you specify in DelPhiController.  Depending on which parameter files  
you use in DelPhi, it might not even be necessary to add hydrogens.

If an amino acid is forming different bonds, it will be protonated  
differently.  If a residue is on the N-terminus of a chain, the most  
likely protonation state (and the one chosen by AddH) is -NH3(+).  If  
a residue is on the C-terminus, the backbone ends in a carboxylate  
with the most likely protonation state -CO2(-).  Force fields, or at  
least Amber, have different sets of charges for the same residue  
depending whether it is on the N-terminus, C-terminus, or in the  
middle of the chain.  If the PDB file says that a cysteine is  
disulfide-bonded to another cysteine (sidechains forming an S-S  
bond), then no hydrogens are put on those sulfurs because they are  
already forming all the bonds they can.  Histidine is a little  
different - the sidechain can be positively charged or neutral  
depending on its environment.  You can have the program decide, or  
control it yourself by changing the residue name as described in the  
man page.  In all these cases the force field will have different  
sets of charges for the different bonding patterns, usually indicated  
by different residue names, that add up to the appropriate formal  
charges.  What these residue names should be depends on the specific  
force field or DelPhi parameter file.

I hope this helps,
Elaine

On Aug 14, 2006, at 8:38 PM, Nihshanka Debroy wrote:

> Hi,
>     I have a couple of questions about the addH utility built into  
> Chimera, and about PDB files in general. I used addH to protonate  
> (add hydrogens) to PDB files. Does it use a force field of some  
> kind in the process, like Amber or Charmm? It seems that depending  
> on the force field used to protonate a PDB, the values for the  
> partial charges and the van der waals radii are different.  
> Specifically, I am using these charges and radii in the  
> DelphiController.
>     Another thing I've noticed is that when adding hydrogens, for  
> the same amino acid, the hydrogens are sometimes different in name  
> or number. For example if I have a Cysteine, in some cases, an HG  
> (gamma hydrogen) is added for the gamma sulphur, balancing out  
> charges completely. In other cases, the HG is not added, and so the  
> amino acid ends up with a net negative charge. I was just wondering  
> if this is supposed to happen.
>    Here's another example. For Alanine, I was getting 1H, 2H and 3H  
> attached to the nitrogen in some cases. In other cases, I just got  
> "H" attached to the nitrogen. Does this mean those particular  
> instances of ALA in the PDB had one only hydrogen attached to the  
> nitrogen, instead of 3 ? Wouldn't the missing 2 hydrogens lead to a  
> net negative charge on the ALA?
>    Thanks for all the help and for this mailing list. It is a  
> wonderful resource.
>
> Sincerely,
> Nihshanka Debroy
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users

-----
Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
UCSF Computer Graphics Lab and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco
                      http://www.cgl.ucsf.edu/home/meng/index.html




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20060815/f3bc2489/attachment.html>