Abstract: The present methods use a variational autoencoder (VAE) and deep generative modelling to learn meaningful representations from the immune repertoires. The system can map input sequences into a lower-dimensional latent space, which reveals a large amount of convergent sequence patterns. The system can identify patterns present in convergent clusters that are highly predictive for antigen exposure and/or antigen specificity. The system can generate, from the latent space, novel functional antibody sequence variants in-silico.

Abstract: The present disclosure describes systems and methods to make predictions classifying one or more properties of a binding protein such as an antibody, for example, antibody affinity or specificity for an antigen. The system can include one or more machine learning models that can extrapolate complex relationships between amino acid sequence and function. The system can be trained on high-quality training data generated through a two-step single-site and combinatorial deep mutational scanning approach. The trained models can then make predictions on novel variant sequences generated in silico. The present disclosure describes amino acid sequences generated by the systems and methods provided, and uses of the generated sequences to produce proteins for therapeutic and diagnostic use.