[Chimera-users] Joining Models + Avoiding Steric Clashes

Eric Pettersen pett at cgl.ucsf.edu
Tue Oct 6 16:22:28 PDT 2015

Hi Carlos,
	I think you are going to have to resort to Python here.  First, look at this page for a basic example of how to loop over some files and do something with them in Chimera/Python:

http://www.cgl.ucsf.edu/chimera/docs/ProgrammersGuide/basicPrimer.html <http://www.cgl.ucsf.edu/chimera/docs/ProgrammersGuide/basicPrimer.html>

	You form the C-N peptide bond using the cnPeptideBond function from the BuildStructure module.  Here’s a pseudocode fragment to form the bond:

	runCommand(“sel atom-spec-to-select-the-N-and-the-C”)
	from chimera.selection import currentAtoms
	n, c = currentAtoms()
	if n.element.name != “N”:
		# the atoms were in the other order, swap them
		n, c = c, n
	from BuildStructure import cnPeptideBond
	cnPeptideBond(c, n, c, 1.33, 180.0, phi=-120.0)

You might want to put the above in yet another loop so that you could try various phi angles.  Naturally you would have to break the bond before trying another value, probably with the ~bond command.
	You might also want to directly measure the clashes in the script rather than write them out to a file.  You do that with the detectClash function in the DetectClash module.  Here’s more pseudocode:

	runCommand(“sel atom-spec to select the C-terminal tail”)
	from chimera.selection import currentAtoms
	from DetectClash import detectClash
	num_clashing_atoms = len(detectClash(currentAtoms()))

Good luck!

	Eric Pettersen
	UCSF Computer Graphics Lab

> On Oct 6, 2015, at 1:49 PM, Carlos Gonzalez Oliver, Mr <carlos.gonzalezoliver at mail.mcgill.ca> wrote:
> Dear Chimera,
> I have been trying to use Join Models to make a C-N bond between the globular domain of a protein and a region of it (a C-terminal tail) whose structure I simulated using MD.
> However, when I do the joining with Join Models C-N bond option, I get some serious steric clashes. I have tried to resolve them by hand by adjusting the bond's torsion and angles but it is too complicated.
> Other than trying to use other software to minimize clashes, I think I might have to use a brute force method and try a bunch of the structures my MD generated until one of them works.
> So what I am trying to do now is write a script to do the model joining automatically with many different structures that I generated, and return the one with the least number of clashes.
> But I haven't been able to find a command line equivalent of the Join Models window.. I was wondering if you know if such a command exists, if not what combination of commands would work? (for repositioning the models and forming the bond)
> Thanks!
> --
> Carlos G. Oliver
> M.Sc. Student
> Vogel Lab
> Department of Biology
> McGill University
> Montréal, QC, Canada
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20151006/c394fd27/attachment.html>

More information about the Chimera-users mailing list