Copy, Move and Merge



In this section we will use the copy, move and merge commands to manipulate the molecular data tree (and implicitly the display data tree). There are several reasons why we may want to reorganize the molecular data tree: For the first operation, moving the atoms of water molecules to level 5, you may use the following command:
Wit!P> build
Build> move -atoms ///hoh*
312 atoms with root at /4apr/A. Target: /4apr/A/W
The atomselection ///hoh* found 312 atoms, with a common root at /4apr/A. The corresponding subtree is then re-linked with a new name (W) into /4apr/A. Now do
Build> list /4apr/A/*
(you must use the keyboard for this, since list is a hidden option in the Build menu, or pick command to recursively call the Wit!P main menu from Build, do your list command, and exit to the build menu again).

You will see that the number of direct descendants of /4apr/A has been reduced from 314 (E, I, HOH_501 ... HOH_885) to 3 (E, I, W). If you Shift-click on a water molecule, you will see that atoms of water molecules now appear at level 5 (e.g. /4apr/A/W/HOH_785/O), and that the residue structure of the water molecules is preserved (i.e. each water molecule is still stored as an individual residue, HOH_501 ... HOH_885).

The water subtree in our example appears as the last direct descendant of /4apr/A. We can exert control over the order of nodes through the -after and -before options of the the move command:

Build> move -atoms ///W
312 atoms with root at /4apr/A/W. Target: -after /4apr/A/E W
The node /4apr/A/W now follows /4apr/A/E at level 3 of the molecular data tree (use list /4apr/A/*, if you want to confirm this). This was just to show you the -after option of the move command. Let's put things back to the previous order, with water molecules at the end. Type
Build> move -atoms /4apr/A/W -done -into /4apr/A W
which will move the subtree defined by the atomselection /4apr/A/W into (as last descendant) /4apr/A, preserving the name W.

Here is how the the molecular data tree looks (down to level 3) after moving the water molecules:

The display tree remains unchanged (c.f. section on molecular data tree).

Next, make a copy of the inhibitor by the command

Build> copy -molecule /4apr/a/I /inh/A/X
which will make a copy of node /4apr/A/I and insert the copy as new node /inh/A/X. This command will also create new nodes in the display tree, so that after the command is completed, the molecular data and display trees will look like this:
 

The new display tree now contains three nodes of type GLBL ("global"), associated with /, /4apr and /inh of the molecular data tree, and two nodes of type MDL ("molecular display list"), one for the /4apr and one for the /inh subtree of the molecular data tree. Since /4apr and /inh are associated with different branches in the display tree, it is now possible rotate and translate them individually, e.g.

Build> display translate /inh 3 3 0
to translate the copy of the inhibitor parallel to the screen by 3 A in x and y (and similarly display rotate ... could be used to apply rotations). Display transformations may be undone by the display reset command:
Build> display reset /inh
causes the copy of the inhibitor to jump back to its original position. Using the display commands with the keyboard is rather clumsy, and of course there is an easier way to do it:
Build> select /inh
selects /inh as a target for the "object" mouse functions (or use the mouse ACR [alt-control-right] function to select the copy, c.f. mouse functions described in the reference section). Simultaneously press the shift key and the middle mouse button, and move the mouse: this will apply a translation to the selected object, /inh. Similarly, you may rotate the selected object by pressing the shift key and the left mouse button while moving the mouse. You may use the
Build> set pivot /inh com /inh
command to set the center of rotation ("pivot") for /inh to the center of mass of inhibitor atoms, or Shift-Alt-Middle Click on an atom to set the pivot to the corresponding atom position.

As an exercise, try docking the copy of the inhibitor into the binding site of 4apr in an upside down binding mode (direction of N-terminal to C-terminal opposite to the N to C direction in the experimental 4apr structure). If you are happy with the result, and want to keep the new binding mode, you will have to turn the display coordinates of the copy of the inhibitor into real coordinates and then merge the inhibitor copy into the 4apr molecular data structure:

Build> freeze /inh
Build> merge /4apr/A /inh/A
The merge command is equivalent to moving all direct descendants of /inh/A into /4apr/A. Without the freeze the new orientation of the copy of the inhibitor would be lost, since the merge command uses real coordinates, not display coordinates. After merging /4apr/A and /inh/A the /inh subtree does not contain any atoms, and is therefore removed from the molecular data tree and from the display tree (a "molecule" without atoms has no right to exist in Wit!P), so that we are left with this:
 


The next section will show you how to prepare the system for minimization.



A.Widmer, NIBR/CPC/CSG-SB