Hi Elo
 
 
The formula for cmputing the putty radius is found in layer1/ Extrude.c
in the source:
 
        switch(source_field) {
        default: /* b*/
          scale = (range+(at->b - mean)/stdev)/range;
          if(scale<0.0F)
            scale = 0.0F;
          scale=(float)pow(scale,power);
          if(scale<min_scale)
            scale=min_scale;
          if(scale>max_scale)
            scale=max_scale;
          *sf = scale;
          break;

>From this I can see:
 
* The "cartoon_putty_scale_max" and "cartoon_putty_scale_min" settings
are not scaling factors, but rather cut-off values to avoid extremely
thin or thick tubes.
* "cartoon_putty_scale_power" is a kind of "gamma correction" which
allows to enhanche areas with extreme b-factors. Set it to 1 if you want
linear scaling to b-factor.
* "cartoon_putty_range" is the only user-adjustable parameter that
affects the actual scaling.  
* The mean value and standard deviation of the b-factors is used for the
scaling. This seems very strange to me - is this a bug or a feature? :-)
 
It is quite hard to make a workaround (without changing the source).
The best I could find was to combine all the (aligned) structures into
one object, but with different chainid, eg
 
alter obj01, chain='A'
alter obj02, chain='B'
create comb, obj01+obj02
 
Then make the putty cartoon
 
show cartoon, comb    # Notice that the complete object must be
selected, the putty scaling works only on the displayed atoms!
cartoon putty, comb
 
But then your putty cartoons of the different structures are - at least
partly - covering each other. Then translate the different chains to
create the view you want, eg:
 
translate [20, 0 0 ], comb and chain A
 
You can use this view directly, or just translate the "unwanted" chains
far enough to be off the screen
 
Hope this helps :-)  
 
But I I would really like an option to do manual scaling on the putty
cartoon, eg by changing the source line to.
 
scale = at->b * cartoon_putty_myscaling; 
 
Where cartoon_putty_myscaling is a new setting.
 
 
Cheers,
 
Esben
 
 
 ----Original Message-----
From: pymol-users-boun...@lists.sourceforge.net
[mailto:pymol-users-boun...@lists.sourceforge.net] On Behalf Of Caspar
Elo Christensen
Sent: 24. juli 2006 22:24
To: PyMOL-users@lists.sourceforge.net
Subject: [PyMOL] Putty



        Hi,
         
        I'm using cartoon putty to display rmsd between an enzyme with
and without ligand. I've got two enzymes (that is four structures all in
all) and have prepared two figures. The problem is that the extreme rmsd
values vary; in one figure it goes from 0.1 to 1.66 while in the other
it goes from 0.1 to 2.33. It seems that the rmsd values are normalized
by putty, so that the figures only allow me to compare the distribution
of rmsd values in each set of structures, not the actual values.How do I
put the two figures on the same scale, so that I can compare them
quantitatively? 
         
        I've tried to play around with the scale range parameters with
little luck.
         
        Any help will be much appreciated,
         
        Elo

Reply via email to