Employing statistical learning to derive species‐level genetic diversity for mammalian species

Abstract

1. The patterns of genetic diversity in several genomic regions have been used in mammalian systematics for decades. For instance, when studying closely related species, it is generally assumed that the mitochondrial cytochrome b gene (cytb) exhibits significant information that can be used for differentiation between intraspecies and interspecies variation in mammals. Because of sampling limitations, early analyses of this proposition were conducted mainly on rodents and bats. 2. Currently, more than 57000 cytb sequences are available covering all major lineages of mammals, and sequencing of several individuals per species is common practice in molecular systematics. We were thus prompted to carry out a large-scale analysis of the utility of cytb genetic variation as a predictor of whether a pair of sequences came from within-species or between-species comparisons. 3. Using predetermined species-level assignments, we employed standard methods from statistical learning to calculate the cutoff values able to classify genetic distances in either intraspecies or interspecies categories; we then measured the performance of such statistical classifiers to predict the species-level taxonomic rank as defined by experts. 4. Depending on the classifier, our results showed that when adopting cytb distance cutoff values of 7.3% and 5.5% for small mammals (Metatheria, Rodentia, Chiroptera, and Eulipotyphla) and 4.3% and 3% for medium-sized to large mammals (Primates, Carnivora, and Artiodactyla), the frequency of incorrect assignment of within-species divergences to the between-species category (type I error) varied from 7 to 11%. 5. In order to avoid over-splitting by future researchers, we calculated cutoff values using a more conservative evaluation and provided a list of mammalian species that are likely to consist of complexes of cryptic species. We show that our supervised method can provide practical guidelines to improve the performance of unsupervised algorithms for species delimitation. Finally, we discuss limitations of large-scale approaches (e.g. effects of misclassification in databases and the need for case-by-case evaluation of cryptic species complexes) and their consequences for conservation policies. SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher's web-site. Appendix S1. Empirically estimated density distributions for within-and between-species genetic distances.

Systematics and diversification of the Ichthyomyini (Cricetidae, Sigmodontinae) revisited: evidence from molecular, morphological, and combined approaches

“…Transient pieces of cartilaginous tissues in young specimens of muroid rodents have been reported before, for example in Peromyscus eremicus ( Spotorno, 1992 ) and plantar-pad polymorphisms have been previously reported for sigmodontines ( e.g ., Musser et al, 1998 : 58). Cytb divergence between specimens from Colombia and Ecuador at 2.8% is well-within reported intraspecific divergence for sigmodontine rodents and other mammals ( Allen et al, 2020 ; Schrago & Mello, 2020 ).…”

Molecular and morphological revision of small Myotinae from the Himalayas shed new light on the poorly known genus Submyotodon (Chiroptera: Vespertilionidae)

“…GenBank has a far greater scope of genetic markers (Porter and Hajibabaei 2018), but because there is currently no possibility to correct or even flag problematic species assignation of sequences, it may actually become increasingly misleading to rely on it for automatized barcoding approaches (Shen et al 2013). For instance, in a recent review evaluating species-level genetic diversity for mammalian species based on CYTB, Schrago and Mello (2020) provided practical guidelines to improve the performance of unsupervised algorithms for species delimitation. To highlight species deserving attention for possible cryptic lineages, they give the example of the endangered M. atacamensis showing "unusually high genetic differences" in that Neotropical species, overlooking the fact that one of the two reference sequences taken from the Genbank (MF143493; Platt et al 2018) is actually a mislabelled M. muricola (see Online Resource 1), hence inflating their "intra-specific" statistics.…”

The Biogeography ofDromiciopsin Southern South America: middle Miocene transgressions, speciation and associations withNothofagus

“…Thus, it is likely that their K2P estimations are low because they included highly conserved nuclear genes in the estimations. Finally, a recent and extensive compilation of more than 57000 mammal cytb sequences including 82 marsupial species showed that 95% of cytb interspecies genetic distances in marsupials fall between 9% and 16% with a mean of 12.3% (Schrago & Mello, 2020). Divergence between D. bozinovici and the other Dromiciops clades falls well within this range.…”

“…Our estimated percentage of divergence in mtDNA between clade A (D. bozinovici, hereafter) and the other clades (range: 7.3-12%) is comparable with what Himes et al (2008) reported (range: 11.3-15%). While these authors did not attribute these divergences to a major split in the Dromiciops lineage, these levels of divergence are typical for inter-specific variation in mammals (Baker & Bradley, 2006;Schrago & Mello, 2020), and led D'Elia et al (2016) to revisit the problem and designing clades A and B to the level of species (Table 5). In contrast, Suárez-Villota et al (2018), analyzing four nuclear and two mtDNA markers in Dromiciops individuals from 20 localities reported genetic distances (K2P) averaged across all loci that were an order of magnitude lower (on the range of 0.001 to 0.015%) and concluded that these differences were within the intra-species range previously reported for other marsupials.…”