TUTORIAL

 

Analysing an active site

 

Building Loops

 

Building a functionnal unit from a monomer

 

Crystal Symmetries

 

Electron Density Maps

 

Energy minimisation

 

Fitting Residues into Electron Density

 

Homology modelling

 

Making Phi/Psi statistics

 

Superposing Proteins

 

 

home


Last modified
31 March 98
by N.Guex

 

Energy minimisation with CHARMm


To complete this tutorial, you need to download some material and install it on the Unix box where you run CHARMm. Then you need to download the pdb file 1APH (bovine insulin) that will be used as an example. As it contains two chains and three SS-bonds, it makes a reasonably complex example.

Step by Step

  • Load the file 1APH.pdb (provided with the tutorial package)

  • The most important thing is to insure that residues of each chain are numbered from 1 for each chain. Otherwise CHARMm will be confused. In this example, everything is fine. But just to be on the safe side (and to learn how to proceed), select all residues of chain A (simply click on one A in the control Panel) and use the "Rename" item of the Edit menu. Type 1 in the appropriate field and repeat the operation for the chain B.

  • Click in the label column of the control panel in front of the Arg1. Go in the display menu and set the label kind to "atom kind". Locate the CZ.

  • Select all residues contained in a sphere of 10Å around the CZ of Arg22. To do so, click on the "display around" icon in the main window (the icon with a little circle). you will be prompted to pick an atom. Pick the Arg22 CZ. A dialog appear. Choose the "Select Groups that are within" radio button and type 10 in the radius field.
    Some residues are selected. To visualise which ones, simply type several times on the enter key of the numeric keypad (PC users should use the + key of the numeric keypad). This will hide and display the selected residues.

  • Use the "reset orientation" item of the edit menu. (The protein might disappear from your field of view do not panic!)

  • Save a CHARMm Energy minimisation job (file menu).

  • The default settings will perform 200 cycles of energy minimisation with the steepest descent method, followed by 200 steps by conjugate gradients, followed by 200 steps by Newton-Rhapson.

  • By default, an harmonic constraint of 2 is added on all CA to prevent them to move far away from their initial location. The higher the constraint the more the atoms will stay around their initial position.

  • To speed-up the tutorial, do only 100 steps of Newton-Rhapson minimisation and accept the settings.

  • Save the file under the name "myjob.pdb"

  • Transmit the file to your Unix computation server in a new directory (the simplest way is to use NFS, but if it is not available on your computer, use ftp).

  • Login on your remote computer and type this command: Cmin myjob.pdb

  • When the energy minimisation is done, you will obtain several files:
    myjob_E.pdb which contains the minimised coordinates and some information about the energy.
    The other files can be useful to trace problems if the minimisation failed. If you don't want them, launch the minimisation with the command "Cmin -r myjob.pdb" and the script will do some clean-up.

  • If you have moved the molecule while the remote computer was computing, your file will not be superposed if you load it directly. Therefore, use the "reset orientation" item of the edit menu immediately before loading the result.

  • Get the file "myjob_E.pdb" by NFS or FTP and load it in the viewer.

  • As you can see, hydrogens capable of H-bonding have been added. If you don't want them, use the appropriate item of the edit menu to discard them.

  • Compare the two proteins and note that only the selected residues have moved. Note that the backbone of those residues has moved. As a supplementary exercise, repeat the same procedure but add a constrain of 0 (zero) to the the backbone in the energy minimisation dialog, and note that only the sidechains of the selected residues have moved.


Known caveats:

Only amino-acids are supported. Nucleotides, solvent and HETATM are ignored but SS-bonds are set-up correctly.

If you want to include some non-standard residues or chemical compounds, you will have to modify the Swiss-PdbViewer minimisation script and provide your own topologies files.