Improved computational epitope profiling using structural models identifies a broader diversity of antibodies that bind to the same epitope

The function of an antibody is intrinsically linked to the epitope it engages. Clonal clustering methods, based on sequence identity, are commonly used group antibodies that will bind same epitope. However, such methods neglect fact with highly diverse sequences can exhibit similar binding site geometries and engage common epitopes. In a previous study, we described SPACE1, method structurally clustered in order predict their This methodology was limited by inaccuracies incomplete coverage template-based modeling. addition, only benchmarked at level domain-consistency one virus class. Here, present SPACE2, which uses latest machine learning-based structure prediction technology combined novel protocol, benchmark data have epitope-level resolution. On six sets antigen-specific antibodies, demonstrate SPACE2 accurately clusters epitopes achieves far higher dataset than clonal SPACE1. Furthermore, show functionally consistent structural identified even more sequence, genetic lineage, species origin those found These results reiterate improve our ability identify epitope, adding information sequence-based especially datasets from sources. openly available GitHub ( https://github.com/oxpig/SPACE2 ).