6. Improving the plots

previous section contents page next section LIGPLOT Operating Manual

6. Improving the plots

The plots produce by LIGPLOT can be improved in one of three ways.

a. Editing the PostScript file

If you are familiar with PostScript files, you can make simple amendments to the plot produced by LIGPLOT by editing the ligplot.ps file.

The file is an ASCII text file, and so can be modified using any text editor. The sorts of amendments you can make are: changes to labels (in terms of size, colour and text), addition of other text, changes to colours, sizes, etc.

Some changes, of course, can be made simply by altering the ligplot.prm parameter file (see section 5) and re-running LIGPLOT.

b. Editing the ligplot.pdb file using interactive computer graphics software

For more radical changes (say to change the positions or orientations of sidechains/residues on the plot), you can use standard interactive computer graphics software, such as QUANTA, to edit the output ligplot.pdb file.

The ligplot.pdb file contains the coordinates of the flattened molecules, exactly as seen on the plot. You can read the file in as a standard PDB file and then use standard molecular modelling operations to modify the structure in any way that will make the final plot clearer.

Of course, the structure will be completely flat and the software might join some non-bonded atoms together by bonds simply because of their proximity to one another. But you can always break the bonds if this will make things clearer.

For example, if you are using QUANTA, you might modify the plot as follows:-

  1. First, import the ligplot.pdb file as a PDB file.

  2. Then, use the "Distance" option of the "Geometry" panel to add dotted lines between hydrogen-bonded atoms. Click first on one, and then the other, of each hydrogen-bonded atom pair. A dotted line will be drawn between the two atoms, showing their distance apart in Ångstroms. This will reproduce the hydrogen bonds shown on the LIGPLOT diagram, and will be useful in getting the distances between them to match the actual distances (printed on the plot) as closely as possible.

  3. Use the "Move Fragment" option of the "Modelling" panel to manually move any of the non-ligand residues around the screen.

    Note that the residues corresponding to hydrophobic contacts will be represented by one (or sometimes more) single carbon atoms. These, too, can be moved around the screen to more favourable positions.

  4. Use the "Torsions" option of the "Modelling" panel to rotate any of the residues or sidechains about any of their rotatable bonds. You can use the torsion angle monitors to ensure that the final torsion angle is always either 0 or 180 degrees (otherwise the molecule will lose its flatness, and subsequent operations may result in greater and greater distortions to the final picture).

  5. Once all the required amendments have been made, save the file as ligplot.pdb, overwriting the previous version. (See note below).

  6. Re-run LIGPLOT, this time using ligplot.pdb as the input file (ie run: ligplot ligplot.pdb [options], where the options defining the ligand are as before). This should produce a new LIGPLOT diagram with all residues laid out as defined on screen.

Note that, when the ligplot.pdb file is saved by QUANTA, any blank chain ID's are replaced by the chain identifier "A". In this case, the residues in the ligplot.pdb file will no longer match the data in the ligplot.hhb, ligplot.nnb and ligplot.bonds files.Thus you will need to edit ligplot.pdb to convert the chain "A" back to chain " " (blank).

Note also, that when LIGPLOT plots the new diagram it uses the information in the ligplot.bonds file to decide where, and of what type, all the bonds are. So it does not matter how many bonds you broke or created in QUANTA, as this information is not stored anywhere.

c. Altering the minimization parameters

The final method of improving a plot is to rerun LIGPLOT, or rather LIGONLY, with different minimization parameters. These can be found at the end of the parameter file, ligplot.prm.

In particular, if the minimization doesn't seem to have been able to run its course, you can change the parameter defining the number of loops for the minimization process.

By upweighting or downweighting certain parameters, you might be able to get closer to the plot you are after - but it is likely to be a very hit-and-miss business, and not one that is recommended.

d. Same protein, different ligands

Sometimes you may want to generate two or more LIGPLOTs for the same protein but with different ligands bound. Direct comparison of such plots can be hampered by the fact that LIGPLOT will probably have arranged the residues in completely different positions on each plot in its efforts to minimise the atom clashes and bond overlaps.

In such cases, it is more useful if the residues on the two (or more) plots are in equivalent positions relative to the ligand. This can be achieved using the ligplot.rcm file which is created whenever you run the program.

The ligplot.rcm file lists the centres of mass (x-y coords) of the residues on the plot (see Appendix B - LIGPLOT file formats). These can be used to restrain the centre-of-mass positions of the residues on a related plot, as follows:

  1. Run LIGPLOT on your first structure and print off the plot produced. Save the ligplot.rcm file generated, as this holds the key positions of the residues on this plot and will form the template for subsequent plots.

      Example - PDB file 8gch (chymotrypsin complex with Gly-Ala-Trp).

    The ligplot.rcm file generated for the above example is: 

                     Res.  Res       Flattened    
            Atom Name  Num      coordinates   
            ---- --- -----    ----------------
RESDUE      CofM GLY C 250      -0.696   3.692
RESDUE      CofM ALA C 251       0.805   0.245
RESDUE      CofM TRP C 252      -3.097  -1.856
RESDUE      CofM HIS -  57       6.414  -3.497
RESDUE      CofM GLY - 216      -4.914   3.844
RESDUE      CofM ASP - 102       6.205  -8.482
RESDUE      CofM SER - 195      -1.501  -8.050
RESDUE      CofM SER - 214       5.391   0.946
RESDUE      CofM TRP - 215      -9.097   1.708
RESDUE      CofM SER - 217      -9.342  -2.440
RESDUE      CofM GLY - 193      -5.669  -6.010
RESDUE      CofM SER - 190     -13.399   0.649
RESDUE      CofM CYS - 191      -8.321  -9.347
RESDUE      CofM GLY - 226      -7.737   6.593
RESDUE      CofM MET - 192     -11.215  -5.617

  2. Run LIGPLOT on your second structure. Identify which residues are present on both plots.

      Example - PDB file 6gch (chymotrypsin complex with n-acetyl-l-phenylalanyl trifluoromethyl ketone).

  3. Copy the ligplot.rcm file, saved in step 1 above, calling it filename.rcm where filename is the name of the PDB file holding your second structure.

  4. Edit filename.rcm to remove any unnecessary residues. You may also need to rename certain residues such that they correspond to the equivalent residues in the second structure. For example, Arg53 in structure 2 might be equivalent to Lys53 in structure 1.

    The above example might now become: 

                     Res.  Res       Flattened    
            Atom Name  Num      coordinates   
            ---- --- -----    ----------------
RESDUE      CofM APF - 246       0.000   0.000
RESDUE      CofM HIS -  57       6.414  -3.497
RESDUE      CofM ASP - 102       6.205  -8.482
RESDUE      CofM SER - 195      -1.501  -8.050
RESDUE      CofM GLY - 193      -5.669  -6.010
RESDUE      CofM CYS - 191      -8.321  -9.347
RESDUE      CofM SER - 190     -13.399   0.649
RESDUE      CofM MET - 192     -11.215  -5.617

  5. Run LIGONLY on your second structure. The residues on the plot should now be restrained to the positions defined by the filename.rcm file, and hence be in roughly the same positions on the page as the residues in the first plot.

      Example - PDB file 6gch with above residue-position restraints.

    Although this gives the residuesin roughly equivalent position, you can see in this example, that the ligand isn't quite in the equivalent orientation, and some of the side-chains of the protein, too, are in different orientations.

    To improve on the plot, you can add atomic restraints. These take the form of ATOM records - which have exactly the same format as ATOM records in a PDB file (see Appendix A - Brookhaven file format).

    Consider, for example, the His57 and Asp102 residues in the plots above. Although these residues are in equivalent positions it would be nice to have them oriented in the same manner. This can be done by removing the ATOM records from the ligplot.pdb file generated by the 8gch plot and inserting them in the 6gch.rcm file as follows:- 

                     Res.  Res       Flattened    
            Atom Name  Num      coordinates   
            ---- --- -----    ----------------
RESDUE      CofM APF - 246       0.000   0.000
ATOM    385  N   HIS    57       7.920  -4.629   0.000  1.00  6.92   6.920
ATOM    386  CA  HIS    57       8.002  -3.175   0.000  1.00  7.11   7.110
ATOM    387  C   HIS    57       9.316  -2.413   0.000  1.00  7.57   7.570
ATOM    388  O   HIS    57       9.345  -1.154   0.000  1.00  8.14   8.140
ATOM    389  CB  HIS    57       6.615  -2.482   0.000  1.00  7.18   7.180
ATOM    390  CG  HIS    57       5.474  -3.458   0.000  1.00  7.22   7.220
ATOM    391  ND1 HIS    57       5.557  -4.819   0.000  1.00  6.95   6.950
ATOM    392  CD2 HIS    57       4.134  -3.163   0.000  1.00  6.92   6.920
ATOM    393  CE1 HIS    57       4.330  -5.324   0.000  1.00  7.49   7.490
ATOM    394  NE2 HIS    57       3.448  -4.350   0.000  1.00  7.23   7.230
ATOM    729  N   ASP   102       3.849  -9.704   0.000  1.00  7.78   7.780
ATOM    730  CA  ASP   102       5.273  -9.283   0.000  1.00  6.95   6.950
ATOM    731  C   ASP   102       6.476 -10.179   0.000  1.00  7.25   7.250
ATOM    732  O   ASP   102       7.641  -9.702   0.000  1.00  7.16   7.160
ATOM    733  CB  ASP   102       5.268  -7.731   0.000  1.00  7.18   7.180
ATOM    734  CG  ASP   102       6.683  -7.212   0.000  1.00  6.68   6.680
ATOM    735  OD1 ASP   102       7.611  -8.064   0.000  1.00  7.10   7.100
ATOM    736  OD2 ASP   102       6.838  -5.976   0.000  1.00  7.75   7.750
RESDUE      CofM SER - 195      -1.501  -8.050
RESDUE      CofM GLY - 193      -5.669  -6.010
RESDUE      CofM CYS - 191      -8.321  -9.347
RESDUE      CofM SER - 190     -13.399   0.649
RESDUE      CofM MET - 192     -11.215  -5.617

      Example - PDB file 6gch with added atom-position restraints for His57 and Asp102.

    This now gets the His57 and Asp102 residues as on the 8gch plot, but the ligand orientation is still not right. So equivalent atom-positions from the 8gch ligplot.pdb file can be taken to fix the ligand. Similarly, the other sidechains can be further tied down with reference to the 8gch plot. The final result might be:

    This now bears a much closer resemblance to the equivalent plot for 8gch. The 6gch.rcm file that gave this plot was:- 

                     Res.  Res       Flattened    
            Atom Name  Num      coordinates   
            ---- --- -----    ----------------
ATOM   1782  F11 APF   246       0.023  -4.549   0.000  0.30 14.18  14.180
ATOM   1784  F13 APF   246      -2.105  -3.859   0.000  1.00 13.79  13.790
ATOM   1788  CP4 APF   246      -6.701  -0.070   0.000  1.00 15.35  15.350
ATOM    385  N   HIS    57       7.920  -4.629   0.000  1.00  6.92   6.920
ATOM    386  CA  HIS    57       8.002  -3.175   0.000  1.00  7.11   7.110
ATOM    387  C   HIS    57       9.316  -2.413   0.000  1.00  7.57   7.570
ATOM    388  O   HIS    57       9.345  -1.154   0.000  1.00  8.14   8.140
ATOM    389  CB  HIS    57       6.615  -2.482   0.000  1.00  7.18   7.180
ATOM    390  CG  HIS    57       5.474  -3.458   0.000  1.00  7.22   7.220
ATOM    391  ND1 HIS    57       5.557  -4.819   0.000  1.00  6.95   6.950
ATOM    392  CD2 HIS    57       4.134  -3.163   0.000  1.00  6.92   6.920
ATOM    393  CE1 HIS    57       4.330  -5.324   0.000  1.00  7.49   7.490
ATOM    394  NE2 HIS    57       3.448  -4.350   0.000  1.00  7.23   7.230
ATOM    729  N   ASP   102       3.849  -9.704   0.000  1.00  7.78   7.780
ATOM    730  CA  ASP   102       5.273  -9.283   0.000  1.00  6.95   6.950
ATOM    731  C   ASP   102       6.476 -10.179   0.000  1.00  7.25   7.250
ATOM    732  O   ASP   102       7.641  -9.702   0.000  1.00  7.16   7.160
ATOM    733  CB  ASP   102       5.268  -7.731   0.000  1.00  7.18   7.180
ATOM    734  CG  ASP   102       6.683  -7.212   0.000  1.00  6.68   6.680
ATOM    735  OD1 ASP   102       7.611  -8.064   0.000  1.00  7.10   7.100
ATOM    736  OD2 ASP   102       6.838  -5.976   0.000  1.00  7.75   7.750
ATOM   1370  N   GLY   193      -3.838  -6.317   0.000  1.00  7.53   7.530
ATOM   1371  CA  GLY   193      -5.074  -5.514   0.000  1.00  6.95   6.950
ATOM   1372  C   GLY   193      -6.327  -6.383   0.000  1.00  6.51   6.510
ATOM   1373  O   GLY   193      -7.437  -5.828   0.000  1.00  6.00   6.000
RESDUE      CofM CYS - 191      -8.321  -9.347
RESDUE      CofM SER - 190     -13.399   0.649
RESDUE      CofM MET - 192     -11.215  -5.617

    Further refinement is still possible, including modifying the coordinates of individual sidechains, say by manipulating them using a graphics program as described in section b above.

previous section contents page next section LIGPLOT Operating Manual