Protein Interaction Interface Analysis Tutorial

Protein-protein interaction interface.Figure 1. Protein-protein interaction interface. (Vagenende V, et al. 2013)

Protein-protein interaction plays an important role in important life processes such as cell fate determination and signal transduction, and is also an important link in the occurrence and development of diseases. Therefore, the protein-protein interaction interface has become an important target for new drug discovery. However, the protein interface has the characteristics of large area of action and relatively flatness, which is not conducive to the binding of drug molecules, especially small-molecule drugs, making the design of drugs based on protein-protein interfaces face severe challenges. Therefore, the development of theoretical calculation methods for protein-white matter interface prediction and corresponding drug design methods for cell signaling pathway research and rational drug design targeting protein-protein interfaces is a very urgent and significant research topic. This article will take influenza A virus H1N1 subtype hemagglutinin/artificially designed HA binding protein binding complex (PDB:3R2X) as an example to introduce how to use the InterfaceAnalyzer in Rosetta to quickly analyze protein-protein interactions and extract valuable Information.

The main applications of InterfaceAnalyzer:

  • Protein-protein molecular docking
  • Protein-protein interaction interface mutation design
  • Protein-protein interaction interface analysis

Note: InterfaceAnalyzer cannot be used for protein-small molecule interaction interface analysis.

1 Prepare the input file

Using Rosetta score_jd2 app to score the input PDB file is a quick way to repair the structure and ensure that the input file can be read correctly by Rosetta.

cd $ROSETTA/demos/public/analyzing_interface_quality/rosetta_inputs

score_jd2.mpi.macosclangrelease -s ../starting_files/*.pdb.gz -no_optH false -ignore_unrecognized_res -out:pdb

mv 3r2x_0001.pdb 3r2x_scored.pdb

Core_jd2 app option note:

-no_optH false: Whether to consider the conformational inversion of glutamine, asparagine and histidine? (Set to True, will help optimize the non-saturated polar atoms embedded in the protein).

-ignore_unrecognized_res: Delete water molecules and other residue molecules that cannot be recognized by Rosetta.

-out:pdb: This will force the score_jd2 application to output a PDB file for later use.

After running, we will get 3r2x_0001.pdb and rename it to 3r2x_scored.pdb as the input file for the next step.

Prepare to run parameter file

Create a new file and enter the following parameters to control the InterfaceAnalyzer calculation mode:

Optimize the mode of the side chain (recommended!)

Non-optimized side chain mode (unless the input structure has been optimized and energy minimized, avoid using it)

Take the parameter file of the optimized side chain mode: pack_input_options.txt file as an example to illustrate the meaning of the parameters.

Run InterfaceAnalyzer

If the protein is a heterodimer composed of A and B chains, we will consider AB as a whole, so we need to add the fixedchains option.

-fixedchains A B stands for: Consider the A chain and the B chain as a whole, without analyzing the interaction between AB.

Open the terminal and enter the following command:

  • InterfaceAnalyzer.mpi.macosclangrelease -s 3r2x_scored.pdb -fixedchains A B @pack_input_options.txt
  • If you need to use a non-optimized side chain mode, replace pack_input_options.txt with no_pack_input_options.txt.
  • If you want the packstat value to be more convergent, you need to add: -packstat::oversample 100

Result analysis

View terminal output: (need to set compute_packstat to true in the parameter file), the scoring results are listed in detail on the screen, at the same time, we can also open the file with a text editor or spreadsheet application to view the analysis results.

Several important indicators in InterfaceAnalyzer:

Protein binding free energy (dG_separated): If you do not use the parameters of the optimized side chain mode, you will find that the calculated ddG value is untrue, because there is an unreasonable arrangement of atom pairs in the input file. Therefore, it is necessary to rearrange the side chains in the calculation process.

The number of unsaturated polar atoms (delta_unsatHbonds): The more unpaired polar atoms, the weaker the protein-protein interaction. It is worth noting that after Rosetta's optimization, more polar atoms with unsaturated bonds may appear, because the program tries to alleviate conflicts between structures.

Packstat: This is an index to measure how well the interface is filled (range 0~1.0). When the value is 1.0, it represents perfect complementarity between proteins. (A value greater than 0.65 is good).

Shape complementarity (shape_complementarity): For antibody analysis, the shape_complementarity score can better represent the complementarity between two molecules.

Embedding surface area (dSASA_int): A measure of the size of the interaction interface, in Angstroms^2.

Binding free energy per unit area (dG_separated / dSASAx100): dG_separated binding energy divided by the embedded surface area value (dSASA_int). (A value below -1.5 is very good).


  1. Vagenende V, Han A X, Pek H B, et al. Quantifying the Molecular Origins of Opposite Solvent Effects on Protein-Protein Interactions. PLoS Computational Biology,9,5(2013-5-16), 2013, 9(5):e1003072.
* For Research Use Only.