Heuristic algorithms for best match graph editing

Abstract Background Best match graphs (BMGs) are a class of colored digraphs that naturally appear in mathematical phylogenetics as representation the pairwise most closely related genes among multiple species. An arc connects gene x with y from another species (vertex color) Y whenever it is one phylogenetically closest relatives . BMGs can be approximated help similarity measures between sequences, albeit not without errors. Empirical estimates thus will usually violate theoretical properties BMGs. The corresponding graph editing problem used to guide error correction for best data. Since set modification problems NP-complete, efficient heuristics needed if practical analysis biological sequence Results have characterization terms consistency certain rooted triples (binary trees on three vertices) defined genes, we consider operate triple sets. As an alternative, show there close connection partitioning leads top-down recursive algorithms similar Aho’s supertree algorithm and give rise BMG consistent sense they leave invariant. Extensive benchmarking shows community detection steps perform editing. Conclusion Noisy data corrected sufficient accuracy efficiency make attractive alternative classical phylogenetic methods.