Command Palette
Search for a command to run...
Paragraph—antibody paratope prediction using graph neural networks with minimal feature vectors
{CM Deane I Moal N Wahome L Chinery}
Abstract
Summary: The development of new vaccines and antibody therapeutics typically takes several years and requires over $1bn in investment. Accurate knowledge of the paratope (antibody binding site) can speed up and reduce the cost of this process by improving our understanding of antibody–antigen binding. We present Paragraph, a structure-based paratope prediction tool that outperforms current state-of-the-art tools using simpler feature vectors and noantigen information. Availability and implementation: Source code is freely available at www.github.com/oxpig/Paragraph. Contact: deane@stats.ox.ac.uk Supplementary information: Supplementary data are available at Bioinformatics online.
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| antibody-antigen-binding-prediction-on-1 | Paragraph | AUC-PR: 0.725 AUC-ROC: 0.934 |
| antibody-antigen-binding-prediction-on-mipe | Paragraph | AUC-PR: 0.650 AUC-ROC: 0.927 |
| antibody-antigen-binding-prediction-on-pecan | Paragraph | AUC-PR: 0.696 AUC-ROC: 0.934 |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.