ABTrans: A Transformer-based model for predicting interaction between anti-Abeta antibodies and peptides

Understanding the recognition of antibodies and A{beta} peptide is crucial for the development of more effective therapeutic agents. Here we studied the interaction between anti-A{beta} antibodies and different peptides by building a deep learning model, using the dodecapeptide sequences elucidated from phage display and known anti-A{beta} antibody sequences collected from public sources. Our multi-classification model, ABTrans was trained to determine the four levels of binding ability between anti-A{beta} antibody and dodecapeptide: not binding, weak binding, medium binding, and strong binding. The accuracy of our model reached 0.8278. Using the ABTrans, we examined the cross-reaction of anti-A{beta} antibodies with other human amyloidogenic proteins, and we found that Aducanumab and Donanemab have the least cross-reactions with other human amyloidogenic proteins. We also systematically screened all human proteins interaction with eleven selected anti-A{beta} antibodies to identify possible peptide fragments that could be an off-target candidate. Key PointsO_LIABTrans is a Transformer-based model that was developed for the first time to predict the interaction between anti-A{beta} antibodies and peptides. C_LIO_LIABTrans was trained using a dataset with 1.5 million peptides and 110 anti-A{beta} antibodies. C_LIO_LIABTrans achieved an accuracy of 0.8278 and is capable of determining the four levels of binding ability between antibody and A{beta}: not binding, weak binding, medium binding, and strong binding. C_LIO_LIABTrans has potential applications in predicting off-target and cross-reactivity effects of antibodies and in designing new anti-A{beta} antibodies. C_LI