RE: [PyMOL] secondary structure and many more questions

[DeLano, W. <wa...@su...>]

That's the spirit Jason -- don't hold back with those questions!

> From: Jason Yano [mailto:jk...@br...]

> 1) Is there a way to find out what secondary strucuture has been=20
> assigned by the automatic algorythem?
> I assume there must be a python command that will dump the=20
> output of the=20
> utility to the screen or to a file.

util.ss
iterate (name ca),print resn,resi,chain,ss

# or to a file (using the persistent global object "stored")

stored.file =3D open("secondary.txt","w")
iterate (name ca),stored.file.write("%s %s %s
%s\n"%(resn,resi,chain,ss))
stored.file.close()

# but remember, that SS algorithm is bogus -- it was cooked up in a few
hours based on what I thought "looked right".=20

> 2) Is there any information available for any of the other=20
> util programs?

The menu module makes use of the util module, so you will see some
example code there.  So far there isn't much else in the way of
documention.=20
=20
> 3) Is it possible to color the chain by B-factor?

# yes, but it's weak.=20

color br0
color br1,(b>5)
color br2,(b>10)
color br3,(b>15)
color br4,(b>20)
color br5,(b>25)
color br6,(b>30)
color br7,(b>35)
color br8,(b>40)
color br9,(b>45)

# You can create a python loop which does the same thing.

# Try colors, s000, s001, ... up through s999 for more subtle shading.
=20
> 4) I have a heme moiety, I can get the program to display the=20
> FE atom,=20
> but I'm not sure how to increase the sphere size of the atom=20
> so that I=20
> can exagerate it, it is possibe? something close to the set=20
> atomradius=20
> command in molscript.

alter (elem fe),vdw =3D 4.0
rebuild

# should work, but just in case you might need to use the
# atom name (whatever it is)

alter (name fe),vdw =3D 4.0
rebuild

> 5) For some strange reason it recognizes the Nitrogens on the heme=20
> differently and colors some of them blue and others pink. I tried to=20
> manually set it like I could for the Fe, but I think that I am not=20
> getting the atom type/name correct. Is there a list of atom names?

PyMOL is probably getting confused.  Please send a copy of the PDB file
to warren@delanoscientific to debug.

> 6) Is there a way to make a C-alpha trace? other than ribbons=20
> something=20
> that will draw straight bonds between C-alpha's?

#ala midas "chain" command?

set ribbon_sampling=3D1
show ribbon

> 7)  Is it possible to change the color of the mesh (not the isomesh)=20
> relatve to helix color ie what are the command line commands. I have=20
> tied color red, mesh and that didn't seem to work, same=20
> question for the=20
> surfaces as well.

set mesh_color=3Dblue
set surface_color=3Dgreen

> 8) Is it possible to have a "ball and stick" representation?

yes, but right now is it kludgy...you will need to create multiple
copies of an object if you want to simultaneously adjust vdw and show
surfaces.

alter (all),vdw=3Dvdw*0.25
set stick_radius=3D0.15
rebuild

> 9) What is the command to output a pov ray file?

You'll need to use Python for this.

cmd.get_povray() returns a tuple of two strings (header and data).
Write these to a  file in order to get a povray input file.
=20
> 10) How do you get two color helicies, ie the equivalent to=20
> planecolour2=20
> in molescript?

sorry can't do this, but 8.5 out of 10 aint bad ; ) PyMOL's secondary
structures are unicolor for now.

> So far the program is running great. The stereo works very=20
> well for me.=20
> I have it running on a dual AMD athlon 1.4GHz, with an Oxygen GVX1=20
> graphics card and 1 gig of  DDR ram under the Windows 2000 pro OS.

I envy you.  That's pretty much the machine I'd like to buy except that
I'd run Linux on it at +200 mhz.  Athlon's are definitely the way to go
-- Python really flies on them.

- Warren