DeepSurf2-FF: Introducing Force Fields in Deep Learning for Improving in-Silico Prediction of Bcr-Antigen Interactions in B Cell Lymphomas

📌 Key Insights:

🔍 Innovation: DeepSurf2-FF integrates force-field data with deep learning to revolutionize BcR-antigen interaction prediction.

📈 Performance: Achieved state-of-the-art results with:

AUC-PR: +17.82%

AUC-ROC: +4.6%

MCC: +5.66%

F-Score: +4.09%

⚡ Impact: Paves the way for more precise therapeutic interventions in B-cell lymphomas.

🧠 Why It Matters:

By leveraging geometric and non-geometric features, DeepSurf2-FF provides unparalleled accuracy in paratope binding site predictions, advancing immuno-hematology research.