Neural message passing for joint paratope-epitope prediction
This work addresses a key challenge in vaccine and synthetic antibody development by improving prediction accuracy for binding sites, though it appears incremental as it builds on prior neural message passing methods.
The paper tackled the problem of predicting antibody-antigen binding sites (paratope and epitope) by proposing distinct neural message passing architectures for each, achieving significant improvements and setting a new state-of-the-art, with qualitative results relevant to COVID-19.
Antibodies are proteins in the immune system which bind to antigens to detect and neutralise them. The binding sites in an antibody-antigen interaction are known as the paratope and epitope, respectively, and the prediction of these regions is key to vaccine and synthetic antibody development. Contrary to prior art, we argue that paratope and epitope predictors require asymmetric treatment, and propose distinct neural message passing architectures that are geared towards the specific aspects of paratope and epitope prediction, respectively. We obtain significant improvements on both tasks, setting the new state-of-the-art and recovering favourable qualitative predictions on antigens of relevance to COVID-19.