Integrative Computational Prediction Strategy for Antibody-Antigen Binding

14 antibody-bound complexes with 9 diverse antigens, including Interleukin-1β, form the test case for a computational strategy developed by Mehmet Emin Aygen. The dataset, last updated in May 2026, supports the evaluation of methods like ESSA and EPIGUIDE for predicting critical binding residues and poses. It includes results from molecular docking and dynamics simulations, with comparisons to AlphaFold3.

Use Cases

• Benchmarking epitope prediction algorithms based on the 14 antibody-antigen complexes.
• Training or validating machine learning models for binding pose prediction using the provided docking results.
• Comparing the performance of coarse-grained methods like ESSA against all-atom simulations for initial screening.
• Guiding experimental validation of antibody-antigen interactions by screening key binding regions.

Strengths

• Includes a challenging test case with Interleukin-1β and its four distinct epitope regions.
• Strategy was applied to 11 diverse antibody-bound complexes with successful results reported.
• Dataset is relatively small at 2.4 MB, facilitating quick download and inspection.

Limitations

• Row count and specific column definitions are unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect methodological bias inherent to the specific computational pipeline described.

Provenance

Source

figshare

Collection Method

Generated via computational strategy involving epitope/paratope prediction, molecular docking (ClusPro), and molecular dynamics simulations.

Freshness

Last updated 2026-05-27 15:06:50; freshness should be verified.