Fig. 3
From: 3GOLD: optimized Levenshtein distance for clustering third-generation sequencing data
SLD cannot accommodate weighted errors. A LD comparison of TAGCTAGC to TAGTAGCT reveals that an insertion of “C’ and a deletion of “T” are required to make the second string match the first for an unweighted edit distance of 2. SLD analysis only considers the insertion of “C” for an unweighted edit distance of 1. B When insertions are weighted 1 and deletions are weighted 5, LD is appropriately 6, reflecting an insertion of “C” and a deletion of “T”. However, SLD is not 1 as expected, but rather 3