Fig. 2.

Two alignments of fragmentation trees based on edge similarities. Nodes represent molecular formulas of the fragments, edges represent molecular formulas of the losses. (a) A gap (−) is introduced for the missing CO loss in the left tree (dashed edge and node). Losses CO and CH₃ are aligned by a mismatch (dotted edges). (b) In the left tree, the fragment after loosing H₃N is missing (dashed edges and node), whereas the fragment after further loss of C₂H₂ is observed. To account for missing fragments, we introduce the join operation. It allows to align the two successive losses H₃N and C₂H₂ in the right tree to a single loss C₂H₅N in the left tree (dotted edges). Fragments may be missing because the corresponding peak was not detected, for example