[Chimera-users] Selection through bond and fragment naming

[Jean Didier Pie Marechal]

Dear all,

thanks a lot for your answers. I seem to have a problem though.
When running the script, one way or another I always ends up with something like:
   a.residue.removeAtom(a)
AttributeError: 'NoneType' object has no attribute 'removeAtom'
AttributeError: 'NoneType' object has no attribute 'removeAtom'

  File "M:\REUNIONS\dibuixos\Manipulate_structure\newres.py", line 20, in <module>
    a.residue.removeAtom(a)

See reply log for Python traceback.

I think I got the origin of the problem but can't figure out how to solve it.

What I observed is that, before running the script if I do the following after selecting my atoms:

>>> for a in at:
	print a.residue
I have the following list printed

	
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1

but doing the same AFTER having run the script gives :

None
UNK 1
UNK 1
UNK 1
UNK 1
UNK 1
...

It seems like the residue type of the first atom disappeared.

Any clue?

all the best,
JD


Dr. Jean-Didier Maréchal
Lecturer
The Computational Biotechnological Chemistry Team
Unitat de Química Física
Departament de Química
Universitat Autònoma de Barcelona
Edifici C.n.
08193 Cerdanyola (Barcelona)
Tel: +34.935814936
e-mail: JeanDidier.Marechal at uab.es
www: http://asklipio.qf.uab.es

----- Missatge original -----
De: Eric Pettersen <pett at cgl.ucsf.edu>
Data: Divendres, Març 11, 2011 8:29 pm
Assumpte: Re: [Chimera-users] Selection through bond and fragment naming

> On Mar 11, 2011, at 7:42 AM, Jean Didier Pie Marechal wrote:
> 
> > Hi,
> >
> > I have a system from which I only have xyz coordinates. What I 
> would  
> > like to do is to define myself residue attribute on different  
> > fragments of the molecules. Could you tell me how (or if there is 
> a  
> > way to ):
> >
> > 1. expand a selection to bonding atoms (i.e. I click on one atom 
> and  
> > expand selection through one bond after each actions) This would 
> be  
> > of the easiest way for me to select the fragments I want to.
> 
> Hi JD,
> 	My only idea as to how to do this is with a short Python script:
> 
> from chimera.selection import currentAtoms, addCurrent
> addAtoms = set()
> for a in currentAtoms():
> 	addAtoms.update(a.neighbors)
> addCurrent(addAtoms)
> 
> 	You could then use the open command to run it:
> 
> open <script folder location (e.g. ~marechal/scripts)>/expand.py
> 
> which you could short to "xp" with the alias command:
> 
> alias ^xp open blah-blah/expand.py
> 
> >
> > 2. set a residue attribute to the given selected fragment. I 
> think I  
> > remember how to do so in scripting, but what is the command line 
> to  
> > do so?
> 
> So, do you really want to set a residue-level attribute, or do you  
> want to organize the selected atoms into a residue?  Setting a 
> residue- 
> level attribute for the atoms is easy:
> 
> 	setattr r myAttrName my-attr-val sel
> 
> Organizing the atoms into a residue again requires Python, and not 
> as  
> simple a script as for your first question, e.g.:
> 
> # script designed to run under "runscript" command, therefore has 
> an  
> "arguments" variable available...
> try:
> 	rname, rnum = arguments
> except ValueError:
> 	raise chimera.UserError("must supply residue name and number  
> arguments")
> 
> try:
> 	rnum = int(rnum)
> except ValueError:
> 	raise chimera.UserError("residue number (second arg) must be an  
> integer")
> 
> atoms = chimera.selection.currentAtoms()
> if not atoms:
> 	raise chimera.UserError("No atoms selected")
> 
> # make a new residue
> newRes = atoms[0].molecule.newResidue(rname, "A", rnum, ' ')
> 
> # withdraw atoms from their current residues
> for a in atoms:
> 	a.residue.removeAtom(a)
> 	if len(a.residue.atoms) == 0:
>        	a.molecule.deleteResidue(a.residue)
> 
> # add atoms to new residue
> for a in atoms:
> 	newRes.addAtom(a)
> 
> 	So you would run this script with the command "runscript blah-
> blah/ 
> newres.py TYR 1", which could again be shorted using aliasing:
> 
> 	alias ^nr runscript blah-blah/newres.py $1 $2
> 
> --Eric
> 
> 
>                         Eric Pettersen
>                         UCSF Computer Graphics Lab
>                         http://www.cgl.ucsf.edu
> 
> 
>