LIGPLOT Operating Manual

---

8. DIMPLOT - interactions across a dimer or domain-domain interface

Written by Alice Wuensche & Roman A Laskowski

The DIMPLOT program produces a plot of the interactions across either a dimer or a domain-domain interface. The interactions plotted include hydrogen bonds and non-bonded contacts. They are extracted from the usual HBPLUS output by the program DIMER; the plot is generated by LIGPLOT, so all the plot parameters are found in the file ligplot.prm and edited in the usual way (see section 5). (In fact, for DIMPLOT, it usually helps to increase the weight of the relative residue-positions energy term in the MINIMIZATION PARAMETERS of the ligplot.prm parameter file).

To run DIMPLOT you need to have installed both the LIGPLOT and HBPLUS programs correctly.

There are two modes of running DIMPLOT: a) for plotting the interactions across a dimer interface, and b) for plotting the interactions across an interface between two domains. These two modes are described below.

a. Running DIMPLOT for a dimer interface

To run DIMPLOT for interactions across a dimer interface, type the following:

dimplot filename chain1 chain1

where:

filename is the name of the PDB file containing your protein structure and should contain the full path unless your structure is in the default directory. The file must be in PDB format (see Appendix A).

chain1 and chain2 are the single-character chain identifiers specifying the two chains defining the interface of interest.

For example, to run DIMPLOT to plot the interface between chains A and B of 3sdp, you might type:

dimplot /data/pdb/p3sdp.pdb A B

A sample output is shown here.

b. Running DIMPLOT for a domain-domain interface

To run DIMPLOT for interactions across a domain-domain interface, type the following:

dimplot filename -d domain1 domain2

where:

filename is the name of the PDB file containing your protein structure and should contain the full path unless your structure is in the default directory. The file must be in PDB format (see Appendix A).

-d indicates that the interface is between two domains within the given PDB file. The domains must be defined in a file with the same name as the PDB file, but with an extension of .dom. The .dom file can either be in the same directory as the PDB file, or in the current working directory. The format of the file must be as defined below.

domain1 and domain2 are two numbers representing the two domains between which the interactions are to be plotted. The domains are as defined in the .dom file (see below).

For example, to run DIMPLOT on domains 2 and 3 of the structure 1atn, you might type:

dimplot /data/pdb/p1atn.pdb -d 2 3

The domain ranges must be in the .dom file, as described below. A sample output is shown here.

c. Format of the domain definitions in the .dom file

If you want to plot the residue interactions across an interface between two domains in a protein you need to define the protein's domains in a .dom file. This file will have the same name as the PDB file, but with an extension of .dom. The file can either be in the same directory as the PDB file, or in the current working directory. The program will look for it in the current directory first.

The file should be an ordinary text file containing lines of the following format:

Domain n: r1 - r2 [chain_id] & r3 - r4 [chain_id] ...

where

n is the domain number (which must be followed by a colon, ":").

r1, r2, r3, etc are residue numbers defining the residue-ranges making up the given domain.

[chain_id] is the chain identifier for the given residue range.

& are the separators between the residue ranges.

The placement of the data is largely free-format, with spaces separating the various data items.

For example, below are the domain definitions for PDB code 1atn (as defined by CATH).

Domain 1:   1 - 136 A & 338 - 371 A 
Domain 2: 137 - 178 A & 272 - 337 A 
Domain 3: 179 - 267 A

This defines domain 1 as consisting of residues 1-136 and 338-371 of chain A, domain 2 consisting of residues 137-178 and 272-337, and domain 3 being residues 179-267.

(Note that, there are two "fragments" not belonging to any domain ranges: residues 268-271 and residue 372).

A sample output for the interface between domains 2 and 3 of 1atn is shown here.

d. DIMONLY - running DIMPLOT on its own

When you run DIMPLOT, as described above, a number of programs are run by the controlling script file. These are: HBADD (see Appendix C), HBPLUS (which calculates all the H-bonds and non-bonded interactions), DIMER (which identifies the interacting residues on either side of the interface) and finally LIGPLOT (which generates the PostScript file).

The whole process can take a long time. If you want a new version of the plot (perhaps after altering the parameters in the ligplot.prm parameter file), it is only necessary to re-run DIMER and LIGPLOT and not any of the earlier programs.

To do this, just type DIMONLY with the same parameters as before. That is:

dimonly /data/pdb/p3sdp.pdb A B

OR

dimonly /data/pdb/p1atn.pdb -d 2 3

---

LIGPLOT Operating Manual