1. Prepare Input File


The AMBER suite consists of 60+ programs.  However, to carried out a traditional molecular dynamics simulation, you only need to know two programs: tleap and sander.  In addition to these two, you will also likely be working with two other helper programs carnal and ptraj.  These two programs are the data analysis workhorse of the AMBER suite, but if you know of other analysis programs that can work with AMBER trajectory files, you can also use those instead. 

Like everything else with computers, running a computer simulation is all about input and output.  sander is the main simulation engine of the entire suite; it takes two input files describing the molecular system to be simulated, one control file specifying the conditions of the simulation and computes a classical molecular dynamics trajectory base on these information.  tleap is the helper program that takes predetermined coordinate files such as pdb and generates a topology file and a restart file.  Of course, I am oversimplifying a bit here.  tleap actually does more than just that and it is not the only program within AMBER that can take a pdb file and generate input files for sander.  The other helper program is called xleap.  The difference between them is that xleap uses graphical interface while tleap uses text-based interface.  When I first started using AMBER, I clung to the graphical xleap and resisted the "backward" text-based tleap.  Over time, I have come to recognize the value of text-based interface.  If you plan to do a lot of simulations, it pays to spend a little extra energy to get used to the text-based interface. 

The latest x-ray structure of the integrase core domain (pdb id 1qs4) has three crystallographic monomers.  There are also several residues missing in the loop region.  Before feeding the pdb file to tleap, we must first fix these problems.  How to model missing residues in a pdb file is highly dependent on the purpose of your study and the particular model you work with.  If you are just getting started with modeling and don't know how to do this sort of things, professor Gale Rhodes at the University of Southern Maine has a very good tutorial on the internet using the freely available SwissPDB Viewer program (http://www.usm.maine.edu/~rhodes/SPVTut/).  In this study, I modeled the missing residues using another crystal structure of the integrase (pdb id 1bis).  In addition to the missing residues, there are also two point mutations at position 133 and 185.  These mutations were introduced mainly to help the protein better form crystals.  However, the F185K mutation was later thought to affect the dynamics of a second surface loop nearby.  Since we are interested in the dynamics of the surface loop, we should change it back from Lys to Phe.  The E133W mutation is not in any critical region of the protein and the pdb file actually says that there is some conflict between the pdb sequence and the corresponding sequence database, so we can just leave the crystal coordinates alone.  These days, mutating residues in the computer is infinitely easier than in the test tube.  There are a number of freely available graphical molecular editing programs that you can use for this.  I used SwissPDB Viewer to do this because it is free and also because it has a built in energy minimization facility to clean up the modeled structure a bit.  To save you some time, you can download the cleaned-up structure from the Download page, or click HERE.

Ok, enough chatter.  Let's get started.

After you've downloaded the file, take a look inside.  You will notice that there are no hydrogen atoms.  If you don't want to use the file I prepared, but want to prepare your own pdb file, be sure to strip all hydrogen atoms now.  Depending on which software you used to build the initial model, you may end up with incompatible hydrogen names.  It is easier to just strip them off and let tleap add the hydrogens. 

On the command line, start up the tleap program.  You should see something like this:

$ tleap
-I: Adding /amber/dat/leap/prep to search path.
-I: Adding /amber/dat/leap/lib to search path.
-I: Adding /amber/dat/leap/parm to search path.
-I: Adding /amber/dat/leap/cmd to search path.

Welcome to LEaP!
Sourcing leaprc: /amber/dat/leap/cmd/leaprc
Log file: ./leap.log
Loading parameters: /amber/dat/leap/parm/parmME.dat
Loading library: /amber/dat/leap/lib/all_nucleic94.lib
Loading library: /amber/dat/leap/lib/all_aminoME.lib
Loading library: /amber/dat/leap/lib/all_aminoctME.lib
Loading library: /amber/dat/leap/lib/all_aminontME.lib
Loading library: /amber/dat/leap/lib/ions94.lib
Loading library: /amber/dat/leap/lib/water.lib
>

Don't worry about the differences in directories.  This can be specified by using your own leaprc file.  You can learn about how to define your leaprc file later.  For now, just accept the default.

If you have problem starting tleap, check to see if your environment variables have been set correctly.  If you are using the bash shell, make sure you have these two lines in your ~/.bash_profile

export AMBERHOME=/amber
export PATH=$PATH:/amber/exe

If you are a C-Shell user, then add the following to your ~/.cshrc file.

setenv AMBERHOME "/amber"
setenv PATH "${PATH}:/amber/exe"

where /amber is the location of your AMBER distribution.

Next, we'll load the pdb file into tleap and assign it to a variable which we will call "mol"

...
...
Loading library: /amber/dat/leap/lib/water.lib
> mol = loadpdb wt1mg.pdb
Loading PDB file: ./wt1mg.pdb
Unknown residue: MG number: 154 type: Terminal/last
..relaxing end constraints to try for a dbase match
-no luck
Added missing heavy atom: .R<CGLN 154>.A<OXT 18>
Creating new UNIT for residue: MG sequence: 155
Created a new atom named: MG within residue: .R<MG 155>
total atoms in file: 1189
Leap added 1192 missing atoms according to residue templates:
1 Heavy
1191 H / lone pairs
The file contained 1 atoms not in residue templates
>
 

If everything went correctly, you should not get any error messages from tleap.  More often than not, you are likely to get some sort of error message the first time you try.  Like here, tleap complains that there is an unknown residue MG.  This is because MG is not included in the parameter libraries we loaded when we started the program.  If your local AMBER installation already has MG parameters included, then you should not see this problem.  If not, you can download a copy of the parameter file from the Download page or simply by clicking HERE.  Quit tleap, place the downloaded file in your current working directory, start tleap again and issue the command "loadoff MG.off" and then loadpdb again.  Your screen should now look like this:

Welcome to LEaP!
Sourcing leaprc: /amber/dat/leap/cmd/leaprc
Log file: ./leap.log
Loading parameters: /amber/dat/leap/parm/parmME.dat
Loading library: /amber/dat/leap/lib/all_nucleic94.lib
Loading library: /amber/dat/leap/lib/all_aminoME.lib
Loading library: /amber/dat/leap/lib/all_aminoctME.lib
Loading library: /amber/dat/leap/lib/all_aminontME.lib
Loading library: /amber/dat/leap/lib/ions94.lib
Loading library: /amber/dat/leap/lib/water.lib
> loadoff MG.off
Loading library: ./MG.off
> mol = loadpdb wt1mg.pdb
Loading PDB file: ./wt1mg.pdb
Added missing heavy atom: .R<CGLN 154>.A<OXT 18>
total atoms in file: 1189
Leap added 1192 missing atoms according to residue templates:
1 Heavy
1191 H / lone pairs
>

The tleap program is smart enough to recognize that the C-terminal are missing an OXT atom, so it automatically added for us.  This is not an error message.

At this point, we have loaded pdb file, added hydrogens and missing heavy atoms, and assigned parameters to all the atoms.  The next step is going to be a bit computationally intensive.  We are going to add a water box and add counter ions to the system to neutralize the charges.


[Previous][Index][Next]