[chimera-dev] hydrogen optimization

Terry Downing terry at francisco.compchem.ucsf.edu
Wed Aug 14 12:58:17 PDT 2002

Eric Pettersen wrote:

> Hi Terry,
>         The AddH extension is pretty simple-minded right now.  The intention for
> the current simple version was that you would take the structure with the
> added hydrogens and save it (in the 1516 release, "pdbrun nouser cat >
> filename" in the midas emulator is the only practical save method; the next
> release will have the midas emulator "write" command implemented).  You
> would then use Amber to minimize the hydrogen positions.
>         Quite a laborious process, I know.
>         I have been working on a smarter AddH that tries to optimize placement
> for hydrogen-bonding potential and minimize steric conflicts, but other
> projects have stolen time from this and I haven't gotten much work done on
> it recently.  Now that I know somebody cares :-), I'll make the smarter
> AddH a higher priority and try to get it into the next release.
>         In the interim, my only suggestion (besides the Amber thing) is that now
> that bond rotations work, you can reduce the steric conflicts by hand by
> rotating the adjacent torsion.  The viability of this approach depends on
> what project you are working on, I suppose.
>         Hope this helps a little, and let us know if you have further thoughts
> or ideas.
> --Eric
> On Tuesday, August 13, 2002, at 06:00 PM, Terry Downing wrote:
> > Hi,
> >
> >   I am attempting to add hydrogens to proteins with ligands bound...I am
> > using the AddH extension, which appears to add hydrogens with reasonable
> > bond lengths and bond angles...However, I notice that because default
> > torsion angles are being used, there are situations in which the
> > cartesian coordinates of the hydrogens are not optimized and result in
> > clashes...This is particularly a problem with hydrogens that are
> > attached to heavy atoms that are free to rotate (ie methyl and amide
> > groups)...Are there any torsion minimization protocols available where I
> > can optimize the torsional angle of just the newly added hydrogens
> > without changing the cartesian coordinates of the heavy atoms (allow
> > them to rotate to optimize the hydrogen placement but not move away from
> > the experimental coordinates)?...I just spoke with Tom Ferrin and he
> > said that Eric was the person most experienced with the AddH extension,
> > and he recommended that I look at the documentation from both him and
> > Elaine...Anything you can recommend on this would be greatly
> > appreciated...
> >
> > Sincerely,
> > P. Therese Downing
> > Kuntz Lab
> >
> > _______________________________________________
> > Chimera-dev mailing list
> > Chimera-dev at cgl.ucsf.edu
> > http://www.cgl.ucsf.edu/mailman/listinfo/chimera-dev

Hey Eric,

  When you put in the minimization, make sure it also includes a torsional
drive...After the hydrogens are added, the programs currently available on the
market minimize the
positions of the hydrogens by setting all of the heavy atoms as fixed, then
minimizing the system...While this will optimize the positions of the hydrogens to
the nearest minima, it will miss potentially better positions since rotational
energy barriers will not be overcome...Thus, a torsional drive is needed to
overcome these barriers and optimize the cartesian location in addition to the bond
lengths and angles...Good luck with the algorithm :O)


More information about the Chimera-dev mailing list