Paragraph-antibody paratope prediction using graph neural networks with minimal feature vectors.

Journal: Bioinformatics (Oxford, England)
Published Date:

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 no antigen information.

Authors

  • Lewis Chinery
    Department of Statistics, University of Oxford, Oxford OX1 3LB, UK.
  • Newton Wahome
    GSK Vaccines, Rockville, MD 20850, USA.
  • Iain Moal
    GSK R&D, Stevenage SG1 2NY, UK.
  • Charlotte M Deane
    Oxford Protein Informatics Group, Department of Statistics, University of Oxford, Oxford, United Kingdom.