A re-analysis of the data in Sharkey et al.’s (2021) minimalist revision reveals that BINs do not deserve names, but BOLD Systems needs a stronger commitment to open science

Abstract

Halting biodiversity decline is one of the most critical challenges for humanity, but biodiversity assessment and monitoring are hampered by taxonomic impediments. We here distinguish between a "dark taxon impediment" caused by a large number of undescribed species and a "superficial description impediment" caused by species descriptions so imprecise that type specimens have to be consulted in order to resolve species identities. Recently, Sharkey et al. (2021) proposed to address the dark taxon impediment for Costa Rican braconid wasps by describing 403 species based on barcode clusters ("BINs") computed by BOLD Systems. The default assumption of the revision is that BIN=Species (e.g., BOLD:ACM9419 becomes Bracon federicomatarritai Sharkey, sp. nov.) and therefore the diagnoses of most species consist only of a consensus barcode. We here argue that this type of "minimalist revision" is unnecessary and undesirable. It is unnecessary because barcode clusters (e.g. BINs) already provide grouping statements that overcome many of the obstacles associated with dark taxon impediments. However, minimalist revisions are also undesirable and problematic because the diagnoses are only based on one character system - that in the case of Sharkey et al. was poorly analyzed. Furthermore, the revision relies on units that violate basic rules of reproducibility because the BINs were delimited by a proprietary algorithm (RESL) that is applied to a mixture of public and private data. Here, we illustrate that many of the BINs described as species are unstable when the available public data are reanalyzed, reiterate that COI mostly measures time of divergence, and that BOLD Systems violates key principles of open science. We conclude by urging authors, reviewers, editors, and grantors to only publish and fund projects that adhere to modern standards of reproducibility.

Species boundaries in the messy middle—A genome‐scale validation of species delimitation in a recently diverged lineage of coastal fog desert lichen fungi

“…In taxonomically challenging groups, the ITS can provide an important first pass for grouping closely related species (Leavitt et al, 2013; Lücking et al, 2014; Moncada et al, 2020). While the taxonomic implications of DNA barcoding approaches remain unsettled (see Meier et al, 2021), our results show that species partitions inferred from the ITS marker coincide with clades comprised of closely related species‐level lineages in Niebla inferred from phylogenomic data (Figure 3). These results suggest that in this speciose genus, the ITS can play a crucial role in assigning specimens to species complexes and provide guidance for future studies characterizing species‐level diversity.…”

Documenting trumpet leaf-miner moths (Tischeriidae): new Neotropical Coptotriche and Astrotischeria species, with notes on Sapindaceae as a host-plant family

“…However, during the last few years, some very controversial attempts rejecting the usual standards for species description have been undertaken to speed species descriptions. Notably, the recent publication "Minimalist revision and description of 403 new species in 11 subfamilies of Costa Rican braconid parasitoid wasps" (Sharkey et al 2021) violates the basic rules of taxonomy (Meier et al 2021).…”

Mangroves are an overlooked hotspot of insect diversity despite low plant diversity

“…However, it is important to remember that barcode-derived units are only proxies for formal species and should only be used for broad analyses of diversity patterns. Additional taxonomic work may uncover, for example, rapid radiations in particular habitats or guilds that would be overlooked by studies that rely on clustering barcodes based on similarity [93]. This is why the rich dolichopodid fauna is currently being revised using the “reverse workflow” approach where morphological validation follows presorting with barcodes [56, 85, 94].…”