Mutations at the interface of a protein-protein complex can alter the binding affinity. In the case of additive multiple mutations, the sum of the changes in binding free energy (ΔΔG_bind) of individual mutations is approximately equal to the ΔΔG_bind value of the complex containing all the mutations. Addem uses machine learning to predict whether a given pair of interface mutations are additive.

• When the user gives a particular PDB ID, Addem checks whether the given structure is that of a heterodimeric complex which does not contain nucleic acid (DNA/RNA).
• If the complex structure is validated, Addem provides a list of interface residues. The user can refer to the JSmol applet for a view of the interface residues, which are displayed in blue.
• Once the user provides the mutations, Addem uses the Random Forest module from scikit-learn (machine learning package from Python) to predict whether the mutations are additive.
• The Result (additive or non-additive) is displayed along with visuals of the complex structure and the mutations.
• If the given pair of mutations are already available in PROXiMATE, the ΔΔG_bind values are displayed for the single and double mutants, and whether it is additive or not.

Flowchart outlining how Addem works

• Give the PDB ID of the protein-protein complex.
• Addem will provide the interface residues in the complex. You can visualize the interface using the JSmol applet.
• Select a pair of interface residues, and then select the mutation for each residue.

Selection and mutation of a pair of interface residues

• Click on the Predict Additivity button.
• You will be directed to a Results page, which will display the complex with the mutations in red.
• The pair of mutations will also be marked "additive" or "non-additive" based on the results of the prediction.
• If the double mutant binding free energy data is available in PROXiMATE, the binding free energy change values will be displayed for the double mutant and its corresponding single mutants.

Sample prediction result for barnase-barstar (1BRS) double mutant A:E73A and D:D35E

What is Addem?

Addem is a prediction server to predict whether the given pair of interface mutations in a given protein-protein complex are additive in terms of free energy of binding. This is NOT a prediction server for binding affinity of mutant protein-protein complexes. To know more about the concept of additivity of binding free energy, please see the questions below.

What are protein-protein interactions?

Nearly every cellular process (cell signalling, cell-cell adhesion, transport, metabolism, ubiquitination etc.) involves protein-protein interactions, i.e. interactions between two or more protein molecules, usually mediated by various biophysical forces (such as electrostatics, Van der Waals forces, hydrogen bonds, hydrophobic interactions etc.). The interaction is characterized by the formation of a protein-protein complex. For detailed reviews of protein-protein interactions, see the following references.

What is additivity of binding free energy?

Additivity in binding affinity of protein-protein complexes refers to the change in free energy of binding (ΔΔG_bind, in kcal/mol) for double (or multiple) mutations which is approximately equal to the sum of their corresponding single mutation ΔΔG_bind values.

Additivity of binding free energy in protein-protein complexes

I am facing issues with the display of the website on my browser.

The website may be best viewed using the latest version of a browser such as IE10+, Google Chrome, or Mozilla Firefox.

I would like to know more about the concept of additivity.

Few studies have been done to study the additivity of binding free energy for mutant protein-protein complexes, for specific complexes or a small dataset of mutations from a few complexes (for e.g., Wells, 1990). Our study of additivity with a dataset of 379 double mutants has revealed significant insights, and the paper is under review.

I have some suggestions/feedback. How to reach you?

Please contact us by filling the form in the Contact page. You will receive a response shortly.