ImaPEp Antibody-Antigen Binding Prediction
ImaPEp predicts binding probabilities for antibody–antigen pairs by representing their binding interfaces as 2D images and leveraging convolutional neural networks.
ImaPEp is a computational tool designed for predicting the binding probability of paratope–epitope pairs in antibody-antigen complexes. Utilizing a novel image-based approach, ImaPEp simplifies 3D structural data into 2D representations of binding interfaces and employs convolutional neural networks to analyze these interactions. By incorporating physicochemical properties and spatial compatibility, ImaPEp achieves high accuracy in identifying binding pairs and offers valuable insights into antibody design. Its efficiency and scalability make it ideal for applications such as docking pose rescoring, high-throughput screening of paratope libraries, and refining antibody-antigen complex structures, thereby accelerating advancements in immunotherapy and biomedical research.
Estimated resource cost: 0.0002945
Categories: Protein Design
Tags: Affinity Prediction Antibodies Developability Proteins
Key Capabilities
- Predicts paratope–epitope binding pairs with high accuracy, achieving a balanced accuracy (BAC) of 0.84 and AUROC of 0.94.
- Introduces a novel 2D image-based approach for antibody–antigen interfaces, leveraging physicochemical properties and spatial compatibility.
- Provides a detailed report, including the number and type of intermolecular contacts, percentage of charged and polar residues, and residue-specific interactions.
Runtime Statistics
| Metric | Value |
|---|---|
| runtime_mean | 7 |
| runtime_median | 7 |
| runtime_std | 1 |
| runtime_90th_percentile | 9 |
| runtime_max | 16 |
Similar Tools
- BindFilter
- BioBind
- ProteinMPNN-ddG
- StaB-ddG
Ready to submit your job?
Review your configuration, then confirm the estimated credit cost before you run the job. Note that credit estimates are not guaranteed and runtime can vary depending on inputs and settings.
Estimated Credits: 0.0002945
Invite-only, limited-time access. Please contact ztang@getantibody.com.
Invite-only, limited-time access. Please contact ztang@getantibody.com.