Supplementary Figure 4 from Long-term Sculpting of the B-cell Repertoire following Cancer Immunotherapy in Patients Treated with Sipuleucel-T

Supplementary Fig. 4: (A-D) BCR convergent groups converge to shared amino acid sequences. (A-B) Phylogenetic trees across all repertories of a representative naive patient (A) and of a representative treated patient (B). For each patient, social network analysis was performed using R packages: Ape (24) and igraph (25). A convergent group was defined as a cluster that includes the clones with the distance less than or equal to 1 (allowing maximum of 1 basepair mutation among clone sequences sharing the same V-gene, J-gene and CDR3 length). The phylogenetic tree was plotted by R package: ggtree (26) with the edge color was coded by amino acid sequences. (C) Social network of a selected V family (V01) across all repertories from the 6 time points of the representative treated patient. Each node represents a single full-length of BCR clonotype sequence colored by its corresponding CDR3 amino acid sequences; node size was characterized based on the corresponding abundance. The nodes connected by lines were within the same convergent group. (D) Social network figures of top 6 largest groups in the V01 family across all repertories from the 6 time points of the representative treated patient in (C). One convergent group was selected for illustration purpose with presenting the actual converged two CDR3 amino acid sequences. (E) Public clusters shared by treated patients at baseline and naïve patients post baseline. Across all repertories from treated patients at baseline and naïve patients post baseline, 5 public clones were identified first. Then social network analysis was performed to find the clones which were at most 1 basepair mismatch with those 5 public clones.