Antibody Affinity Optimization¶
Antibody affinity optimization aims to improve the binding affinity between an antibody and its target antigen. It is a critical step in antibody engineering.
Features¶
GearBind uses advanced deep learning models to predict the impacts of mutations on antibody-antigen binding affinity. It allows users to quickly evaluate a large number of antibody variants in silico and identify the most promising ones for experimental validation.
The mutation sites can be specified flexibly. For example, you may want to target the CDR regions or all residues that contact the antigen. GearBind will return the top mutations ranked by predicted binding affinity improvements.
Inputs¶
To submit an Antibody Affinity Optimization job, open the Project Editor and select "Affinity Optimization" from the "Binding Analysis" dropdown menu.
- Antibody-Antigen Complex: Upload a PDB or mmCIF file containing the antibody-antigen complex structure.
- Ligand: Specify the chain ID(s) of the ligand (antibody), e.g. "H,L".
- Target: Specify the chain ID(s) of the target (antigen), e.g. "C".
- Mutation Sites: Residues to be mutated, in the format of comma-seperated "chain:start-end", e.g. "H:100-112,L:50-60". If the complex is currently open in the Structure Viewer, you may click on "import from selection" to fill in this input box automatically.
- Top Mutations: The number of top mutations to return. Must not exceed the product of the number of mutation sites and the number of amino acid types.
- Job Name: Name of the job. Note that the job name must be unique within the project.
Results¶
The results contain the predicted binding affinity changes (DDG) of the mutations and the modeled mutant structures.
Click the job name in the Files & Jobs panel to view the results.
The main results are summarized in a table (summary.csv) with columns:
- mutation: Mutation in the format of "chain+residueID+mutAA", e.g. "H103Y".
- ddg: Predicted binding free energy change of the mutation in kcal/mol. More negative value indicates stronger binding.
- file_urls: URLs to the modeled mutant PDB files.
You may click the "file" icon to visualize and analyze the mutant structures.