Fig. 1.

Partitioning a CDR3 sequence: boxes correspond to nucleotides and are indexed by integers. Each group of three boxes (identified by heavier boundary lines) corresponds to an amino acid. The nucleotide positions $x_1,\dots ,x_4$ identify the boundaries between different elements of the partition. The $\mathcal{V},\hspace{0.17em}\mathcal{M},\hspace{0.17em}\mathcal{D}(D),\hspace{0.17em}\mathcal{N}$ and $\mathcal{J}(D)$ matrices define cumulated weights corresponding to each of the five elements