Statistical Potential to Improve Antibody-Antigen Docking

In this paper, we describe the development of a novel statistical potential for the prediction of antibody-antigen complexes (docking), which play key role in in silico immunotherapy discovery. The developed statistical potential is then used to improve the accuracy of an existing docking algorithm. We also present a new dataset for the development and comparison of different statistical potentials and docking algorithms. One of the key features of the developed dataset is that it can be obtained almost automatically, with few optional manual steps, using the pipeline introduced in this paper