LigPlot in detail.

 A few weeks back the P212121 blog posted a note about Ligplot (you can read it here). I realized that I wasn't very much aware of any updates to Ligplot or its dependencies (Hbplus and Naccess).

So, I went about to renew my Ligplot license and get it going. While doing this I noticed another file, the Het Group Dictionary (found here http://pdb.rutgers.edu/het_dictionary.txt), that also got updated.
With all this files in hand I compiled as indicated in the installation document and place in my usual binary directory, /sw64/bin/custom/. Fink users will notice the familiar file structure.

Once there is setup a directory with a single PDB, in this case, 1LUQ. I double check its quality at the Molprobity website and used the final PBD file without protons.

Ligplot is run form the command line like so:

Once this is properly done you'll get quite a few files, the most relevant will mostly be ligplot.ps. For my example, this file looks like so (you can click on it to see a larger version):

Biotin bound to streptavidin

This is pretty much the default output but Ligplot can read accessibility output produced by Naccess. Besides running Naccess before Ligplot you have to modify a file called ligplot.prm:

All you have to do is change the accessibility shading from N to Y and you are set:

An interesting side-effect file being created from running Ligplot is the one called ligplot.frm. This is a PDB file with the coordinates of the residues and ligand shown in the images above (hydrogens have been removed for clarity):

This is fairly useful since you can use this file with the original just to have a predefined selection for the ligand environment. It also allows you to "see" how the hydrophobic contacts are contributing to binding.

Now, for comparison, here is a sample of a similar image created in PyMOL:

Generated by manually selecting BTN from chain B then expand the selection by 5 A for residues and, finally, generating porlar interactions within selection.

You get pretty much the same thing. However, the PyMOL generated image took several clicks and it includes polar interactions within the binding residues. It is a bit harder to distinguish them from the ones to the ligands.

Ok, another example is this: imagine you have this huge protein, 500 kDa lots and lots of residues. For whatever reason you want to not only locate a particular residue but also to get a good look at its environment. In the following example I picked a large protein that has only one tryptophan. I gave its residue number and chain to ligplot and:

Ok, good. But then, there is the ligplot.frm file:

Pretty useful, don't you think.

Botton line is, Ligplot is very useful to check/review/analysis binding sites. It is not unique for 3D but it is great for 2D too. The nice 2D images of said binding sites can be used for offscreen reference and a 3D one that you can further use to work with in any program of your choice. Ligplot can be used to review a residue environment, not only ligands.

I recomend it!