The AVI (5 Mb) |
The MPEG (1.2 Mb) |
An small animated Gif (2.8 Mb) |
B.S. Kang, D.R. Cooper, Y. Devedjiev, U. Derewenda, Z.S. Derewenda. Molecular roots of degenerate specificity in syntenin's PDZ2 domain: reassessment of the PDZ recognition paradigm. Structure. In press.This last structure was created by pasting the two peptides into the unbound structure and then moving the peptides so they would be out of frame when the movie starts. It looked like this.

I used LSQMAN to morph between the unbound and both bound forms. Since you might want to do this multiple times (ie with different start positions of the peptides) it is best to do this with a script.
NOTE: LSQMAN renames your chains A->Z.
Here is the script I used to do one of the morphings.
lx_lsqman <<end-lsq #Read in the two files re m1 2pdz-w.pdb re m2 spep-w.pdb #Tell LSQMAN to use all atoms at no #Fix silly nomenclature problems nomen m1 nomen m2 fix m1 a197-270 b5-8 c2-6 m2 a197-270 b5-8 c2-6 strict seq torsion #Morph between the two structures morph m1 A197-270 C2-6 m2 A197-270 C2-6 30 morpha c m A197-270 999 quit end-lsq
This script will generate morpha_1.pdb -> morpha_30.pdb
ls morph?_?.pdb | awk -F "_" '{print ("mv "$0" " $1"_0"$2)}' | /bin/sh
The Pymol manual says not to trust its secondary structure assignment procedure (which is slow anyway). I used DSSP and a helper script dssp2pdb to add the secondary structure to each intermediate files. I used the following script to add the secondary structure elements to all the morphed PDB files.
#!/bin/bash #Run dssp on the starting structure dssp 2pdz-w.pdb 2pdz-w.dssp #Get the stuff to add to each file dssp2pdb -35 2pdz-w.dssp > tEmP #If a morphed pdb doesn't already have it, add the secondary structure info. for i in `ls morph*pdb`; do if ! [ `grep -l HELIX $i` ]; then cat tEmP $i > tEmP2 mv tEmP2 $i fi done rm -f tEmP
In Pymol, each morphed structure is considered one state of a molecule. Each state is loaded with the command
load file,object,state#
To automatically generate the "loading" part of the pymol script, I used the following command:
ls morph*pdb |awk '{print ("load",$0",mov,"NR)}' > load.pml
which loads each PDB into a different state of the molecule mol with these
commands.
load morpha_01.pdb,mov,1 load morpha_02.pdb,mov,2 ... load morphb_29.pdb,mov,59 load morphb_30.pdb,mov,60
Then generate the command that will do the movie. In this case:
mset 1 -30 30 x15 30 -1 31 -60 60 x15 60 -31Which in order
set_view (\
-0.534639537, -0.058760803, 0.843031168,\
0.793129861, -0.379265100, 0.476552665,\
0.291729540, 0.923420310, 0.249373451,\
-3.371663570, 2.114262581, -87.563941956,\
17.698165894, 8.629757881, 12.076782227,\
43.687160492, 144.238052368, 0.000000000 )
select synt, ( chain a )
select spep, (chain c)
select ipep, (chain b)
select pocket, (chain a & resi 207:214 | resi 250:260)
select pocket2, (chain a & resi 208:213 | resi 251:259)
color grey, (synt & ! pocket)
show cartoon, (synt)
hide lines, (synt)
hide cartoon, (pocket2)
util.cbay (pocket)
show sticks, (pocket)
show sticks, (spep or ipep)
util.cbac (spep)
util.cbak (ipep)
set ambient = .13
set gamma = 2.0
select none
The background was a GCO (Compiled Graphics Object). It consists of almost perpendiculat triangels. One triangle is the floor and the other is the back wall. You will have to fiddle with the coordinates of the verticies to get this to look right.
Here is a side view so you can see the two triangles clearly.

# floor.py from pymol.cgo import * from pymol import cmd from pymol.vfont import plain obj = [ BEGIN, TRIANGLES, COLOR, 1,1,1, VERTEX, 25,-30,-10, VERTEX, -30,30,11, VERTEX, 50,50,-5, COLOR, .0, .0, 1, VERTEX, 25,-30,-10, VERTEX, -30,30,11, VERTEX, -10,-10,100, END ] # then we load it into PyMOL cmd.load_cgo(obj,'floor')It is loaded into pymol with the command
run floor.py
Here is the whole script, load.pml
@load.pml set ray_trace_frames = 1 mpng peptide-frameThis creates individual frames in the PNG format.
If anyone knows of a good way to do this in Linux, please let me know. Convert can be used to make animated GIFs and MPEGS (with pretty bad resolution--maybe it could be tweeked, so if you know how let me know.
I used a Windows Shareware program Video Mach to make an AVI. The AVI file has descent resolution and plays without a hitch in Powerpoint.