Figure 2.

Incomplete breakpoint graph and missing adjacencies inference. (A) Two reference genomes (R₁ and R₂) and a target genome T represented in the alphabet of four signed synteny blocks (a, b, c, d). All genomes are circular. Reference R₁ could be transformed into R₂ with two inversions: (a, b, c, d) → (a, −c, −b, d) → (a, −c, −d, +b). Target genome T is structurally similar to R₁; however, T is fragmented into three contigs (missing adjacencies are shown with dashed lines). (B) Incomplete breakpoint graph of all genomes. Each synteny block corresponds to two nodes representing its head and tail (denoted as h and t). Colored edges connect synteny block ends that are adjacent within a corresponding genome. Two inversions between R₁ and R₂ correspond to two cycles of length four with alternating colors. Because T is fragmented, some red adjacencies are missing (dashed lines). (C) A phylogenetic tree is reconstructed based on pairwise breakpoint distance between the genomes. Because T shares more breakpoints with R₁ than with R₂, T is closer to R₁ within the tree. (D) Missing adjacencies are recovered based on the graph structure and phylogenetic tree. Since T is closer to R₁, the algorithm prefers the pair of breakpoints (b^t, c^h), (d^t, a^h) to an alternative, (b^t, a^h), (c^h, d^t). The reconstructed adjacencies define the synteny block order in the final chromosomes.