FIGURE 3.

Results of agglomerative clustering as a function of RMSD threshold on the Cao et al. (2023) training set. Agglomerative clustering with a “complete” linkage criterion was performed for threshold values between 0.5 and 5 Å. The values of the five performance metrics are plotted against evaluated threshold values: (A) fraction of epitope-consistent clusters, (B) fraction of clustered antibodies in epitope-consistent clusters, (C) number of multiple-occupancy clusters, (D) number of antibodies in multiple-occupancy clusters, and (E) number of antibodies in epitope-consistent multiple-occupancy clusters. Results of a random clustering baseline (see Materials and Methods) are shown for comparison. Values for the number of multiple-occupancy clusters and antibodies in multiple-occupancy clusters for the random baseline are matched to agglomerative clustering.