Revisiting use of DNA characters in taxonomy with MolD - a tree independent algorithm to retrieve diagnostic nucleotide characters from monolocus datasets

While DNA characters are increasingly used for phylogenetic inference, taxa delimitation and identification, their use for formal description of taxa (i.e. providing either a formal description or a diagnosis) remains scarce and inconsistent. The impediments are neither nomenclatural, nor conceptual, but rather methodological issues: lack of agreement of what DNA character should be provided, and lack of a suitable operational algorithm to identify such characters. Furthermore, the reluctance of using DNA data in taxonomy may also be due to the concerns of insufficient reliability of DNA characters as robustness of the DNA based diagnoses has never been thoroughly assessed. Removing these impediments will enhance integrity of systematics, and will enable efficient treatment of traditionally problematic cases, such as for example, cryptic species. We have developed a novel versatile and scalable algorithm MolD to recover diagnostic combinations of nucleotides (DNCs) for pre-defined groups of DNA sequences, corresponding to taxa. We applied MolD to four published monolocus datasets to examine 1) which type of DNA characters compilation allows for more robust diagnosis, and 2) how the robustness of DNA based diagnosis changes depending on the sampled fraction of taxons diversity. We demonstrate that the redundant DNCs, termed herein sDNCs, allow for higher robustness. Furthermore, we show that a reliable DNA-based diagnosis may be obtained when a rather small fraction of the entire data set is available. Based on our results we propose improvements to the existing practices of handling DNA data in taxonomic descriptions, and discuss a workflow of contemporary systematic study, where the integrative taxonomy part precedes the proposition of a DNA based diagnosis and the diagnosis itself can be efficiently used as a DNA barcode. Our analysis fills existing methodological gaps, thus setting stage for a wider use of the DNA data in taxa description.