Using natural history to guide supervised machine learning for cryptic species delimitation with genetic data

Cryptic species conservation: a review

Cryptic species are groups of two or more taxa that were previously classified as single nominal species. Being almost morphologically indistinguishable, cryptic species have historically been hard to detect. Only through modern morphometric, genetic, and molecular analyses has the hidden biodiversity of cryptic species complexes been revealed. Cryptic diversity is now widely acknowledged, but unlike more recognisable, charismatic species, scientists face additional challenges when studying cryptic taxa and protecting their wild populations. Demographical and ecological data are vital to facilitate and inform successful conservation actions, particularly at the individual species level, yet this information is lacking for many cryptic species due to their recent taxonomic description and lack of research attention. The first part of this article summarises cryptic speciation and diversity, and explores the numerous barriers and considerations that conservation biologists must navigate to detect, study and manage cryptic species populations effectively. The second part of the article seeks to address how we can overcome the challenges associated with efficiently and non-invasively detecting cryptic species in-situ, and filling vital knowledge gaps that are currently inhibiting applied conservation. The final section discusses future directions, and suggests that large-scale, holistic, and collaborative approaches that build upon successful existing applications will be vital for cryptic species conservation. This article also acknowledges that sufficient data to implement effective species-specific conservation will be difficult to attain for many cryptic animals, and protected area networks will be vital for their conservation in the short term.

Species delimitation 4.0: integrative taxonomy meets artificial intelligence

Although species are central units for biological research, recent findings in genomics are raising awareness that what we call species can be ill-founded entities due to solely morphology-based, regional species descriptions. This particularly applies to groups characterized by intricate evolutionary processes such as hybridization, polyploidy, or asexuality. Here, challenges of current integrative taxonomy (genetics/genomics + morphology + ecology, etc.) become apparent: different favored species concepts, lack of universal characters/markers, missing appropriate analytical tools for intricate evolutionary processes, and highly subjective ranking and fusion of datasets. Now, integrative taxonomy combined with artificial intelligence under a unified species concept can enable automated feature learning and data integration, and thus reduce subjectivity in species delimitation. This approach will likely accelerate revising and unraveling eukaryotic biodiversity.

Establishing species boundaries in Bornean geckos

Species delimitation using mitochondrial DNA (mtDNA) remains an important and accessible approach for discovering and delimiting species. However, delimiting species with a single locus (e.g. DNA barcoding) is biased towards overestimating species diversity. The highly diverse gecko genus Cyrtodactylus is one such group where delimitation using mtDNA remains the paradigm. In this study, we use genomic data to test putative species boundaries established using mtDNA within three recognized species of Cyrtodactylus on the island of Borneo. We predict that multi-locus genomic data will estimate fewer species than mtDNA, which could have important ramifications for the species diversity within the genus. We aim to (i) investigate the correspondence between species delimitations using mtDNA and genomic data, (ii) infer species trees for each target species, and (iii) quantify gene flow and identify migration patterns to assess population connectivity. We find that species diversity is overestimated and that species boundaries differ between mtDNA and nuclear data. This underscores the value of using genomic data to reassess mtDNA-based species delimitations for taxa lacking clear species boundaries. We expect the number of recognized species within Cyrtodactylus to continue increasing, but, when possible, genomic data should be included to inform more accurate species boundaries.

Machine learning approaches delimit cryptic taxa in a previously intractable species complex

Cryptic species are not diagnosable via morphological criteria, but can be detected through analysis of DNA sequences. A number of methods have been developed for identifying species based on genetic data; however, these methods are prone to over-splitting taxa with extreme population structure, such as dispersal-limited organisms. Machine learning methodologies have the potential to overcome this challenge. Here, we apply such approaches, using a large dataset generated through hybrid target enrichment of ultraconserved elements (UCEs). Our study taxon is the Aoraki denticulata species complex, a lineage of extremely low-dispersal arachnids endemic to the South Island of Aotearoa New Zealand. This group of mite harvesters has been the subject of previous species delimitation studies using smaller datasets generated through Sanger sequencing and analytical approaches that rely on multispecies coalescent models and barcoding gap discovery. Those analyses yielded a number of putative cryptic species that seems unrealistic and extreme, based on what we know about species' geographic ranges and genetic diversity in non-cryptic mite harvesters. We find that machine learning approaches, on the other hand, identify cryptic species with geographic ranges that are similar to those seen in other morphologically diagnosable mite harvesters in Aotearoa New Zealand's South Island. We performed both unsupervised and supervised machine learning analyses, the latter with training data drawn either from animals broadly (vagile and non-vagile) or from a custom training dataset from dispersal-limited harvesters. We conclude that applying machine learning approaches to the analysis of UCE-derived genetic data is an effective method for delimiting species in complexes of low-vagility cryptic species, and that the incorporation of training data from biologically relevant analogues can be critically informative.

Morphological Species Delimitation in The Western Pond Turtle (Actinemys): Can Machine Learning Methods Aid in Cryptic Species Identification?

As the discovery of cryptic species has increased in frequency, there has been an interest in whether geometric morphometric data can detect fine-scale patterns of variation that can be used to morphologically diagnose such species. We used a combination of geometric morphometric data and an ensemble of five supervised machine learning methods (MLMs) to investigate whether plastron shape can differentiate two putative cryptic turtle species, Actinemys marmorata and Actinemys pallida. Actinemys has been the focus of considerable research due to its biogeographic distribution and conservation status. Despite this work, reliable morphological diagnoses for its two species are still lacking. We validated our approach on two datasets, one consisting of eight morphologically disparate emydid species, the other consisting of two subspecies of Trachemys (T. scripta scripta, T. scripta elegans). The validation tests returned near-perfect classification rates, demonstrating that plastron shape is an effective means for distinguishing taxonomic groups of emydids via MLMs. In contrast, the same methods did not return high classification rates for a set of alternative phylogeographic and morphological binning schemes in Actinemys. All classification hypotheses performed poorly relative to the validation datasets and no single hypothesis was unequivocally supported for Actinemys. Two hypotheses had machine learning performance that was marginally better than our remaining hypotheses. In both cases, those hypotheses favored a two-species split between A. marmorata and A. pallida specimens, lending tentative morphological support to the hypothesis of two Actinemys species. However, the machine learning results also underscore that Actinemys as a whole has lower levels of plastral variation than other turtles within Emydidae, but the reason for this morphological conservatism is unclear.

Unsupervised machine learning for species delimitation, integrative taxonomy, and biodiversity conservation

Integrative taxonomy, combining data from multiple axes of biologically relevant variation, is a major goal of systematics. Ideally, such taxonomies will derive from similarly integrative species-delimitation analyses. Yet, most current methods rely solely or primarily on molecular data, with other layers often incorporated only in a post hoc qualitative or comparative manner. A major limitation is the difficulty of devising quantitative parametric models linking different datasets in a unified ecological and evolutionary framework. Machine Learning (ML) methods offer flexibility in this arena by easily learning high-dimensional associations between observations (e.g., individual specimens) across a wide array of input features (e.g., genetics, geography, environment, and phenotype) to delimit statistically meaningful clusters. Here, I implement an unsupervised method using Self-Organizing (or "Kohonen") Maps (SOMs) for such purposes. Recent extensions called "SuperSOMs" can integrate multiple layers, each of which exerts independent influence on a two-dimensional output grid via empirically estimated weights. The grid cells are then delimited into K distinct units that can be interpreted as species or other entities. I show empirical examples in salamanders (Desmognathus) and snakes (Storeria) with layers representing alleles, space, climate, and traits. Simulations reveal that the SuperSOM approach can detect K = 1, tends not to over-split, reflects contributions from all layers, and limits large layers (e.g., genetic matrices) from overwhelming other datasets, desirable properties addressing major concerns from previous studies. Finally, I suggest that these and similar methods could integrate conservation-relevant layers such as population trends and human encroachment to delimit management units from an explicitly quantitative framework grounded in the ecology and evolution of species limits and boundaries.

An Opiliones-specific ultraconserved element probe set with a near-complete family-level phylogeny

Sequence capture of ultraconserved elements (UCEs) has transformed molecular systematics across many taxa, with arachnids being no exception. The probe set available for Arachnida has been repeatedly used across multiple arachnid lineages and taxonomic levels, however more specific probe sets for spiders have demonstrated that more UCEs can be recovered with higher probe specificity. In this study, we develop an Opiliones-specific UCE probe set targeting 1915 UCEs using a combination of probes designed from genomes and transcriptomes, as well as the most useful probes from the Arachnida probe set. We demonstrate the effectiveness of this probe set across Opiliones with the most complete family-level phylogeny made to date, including representatives from 61 of 63 currently described families. We also test UCE recovery from historical specimens with degraded DNA, examine population-level data sets, and assess "backwards compatibility" with samples hybridized with the Arachnida probe set. The resulting phylogenies - which include specimens hybridized using both the Opiliones and Arachnida probe sets, historical specimens, and transcriptomes - are largely congruent with previous multi-locus and phylogenomic analyses. The probe set is also "backwards compatible", increasing the number of loci obtained in samples previously hybridized with the Arachnida probe set, and shows high utility down to shallow population-level divergences. This probe set has the potential to further transform Opiliones molecular systematics, resolving many long-standing taxonomic issues plaguing this lineage.

Taxonomic Delineation of the Old World Species Stomphastis thraustica (Lepidoptera: Gracillariidae) Feeding on Jatropha gossypiifolia (Euphorbiaceae) that Was Collected in the New World and Imported as a Biocontrol Agent to Australia

We provide the identification and species delineation of this biocontrol agent as Stomphastis thraustica (Meyrick in Trans Ent Soc Lond 80(1):107-120, 1908) belonging to the family Gracillariidae. We clarify the distribution pattern of S. thraustica, its host plant preferences, and present taxonomic and molecular diagnoses based on original morphological and genetic data as well as data retrieved from historic literature and genetic databases. Following our own collecting efforts in three continents Africa, South America, and Australia as well as our study of historic museum collection material, we present many new distribution records of S. thraustica for countries and territories in the world including the new discovery of this species in the Neotropical region and we report its introduction in Australia as a biocontrol agent. Using mitogenomic and COI gene data, we clarified that the closest relative of S. thraustica is Stomphastis sp. that occurs in Madagascar and Australia and feeds on the same host plant as S. thraustica - Jatropha gossypiifolia L. (Euphorbiaceae). The molecular sequence divergence in the mitochondrial DNA barcode fragment between these two closely related species S. thraustica and Stomphastis sp. is over 5.7% supporting that they are different species.

From genomics to integrative species delimitation? The case study of the Indo-Pacific Pocillopora corals

With the advent of genomics, sequencing thousands of loci from hundreds of individuals now appears feasible at reasonable costs, allowing complex phylogenies to be resolved. This is particularly relevant for cnidarians, for which insufficient data is available due to the small number of currently available markers and obscures species boundaries. Difficulties in inferring gene trees and morphological incongruences further blur the study and conservation of these organisms. Yet, can genomics alone be used to delimit species? Here, focusing on the coral genus Pocillopora, whose colonies play key roles in Indo-Pacific reef ecosystems but have challenged taxonomists for decades, we explored and discussed the usefulness of multiple criteria (genetics, morphology, biogeography and symbiosis ecology) to delimit species of this genus. Phylogenetic inferences, clustering approaches and species delimitation methods based on genome-wide single-nucleotide polymorphisms (SNP) were first used to resolve Pocillopora phylogeny and propose genomic species hypotheses from 356 colonies sampled across the Indo-Pacific (western Indian Ocean, tropical southwestern Pacific and south-east Polynesia). These species hypotheses were then compared to other lines of evidence based on genetic, morphology, biogeography and symbiont associations. Out of 21 species hypotheses delimited by genomics, 13 were strongly supported by all approaches, while six could represent either undescribed species or nominal species that have been synonymised incorrectly. Altogether, our results support (1) the obsolescence of macromorphology (i.e., overall colony and branches shape) but the relevance of micromorphology (i.e., corallite structures) to refine Pocillopora species boundaries, (2) the relevance of the mtORF (coupled with other markers in some cases) as a diagnostic marker of most species, (3) the requirement of molecular identification when species identity of colonies is absolutely necessary to interpret results, as morphology can blur species identification in the field, and (4) the need for a taxonomic revision of the genus Pocillopora. These results give new insights into the usefulness of multiple criteria for resolving Pocillopora, and more widely, scleractinian species boundaries, and will ultimately contribute to the taxonomic revision of this genus and the conservation of its species.

New species in old mountains: integrative taxonomy reveals ten new species and extensive short-range endemism in Nesticus spiders (Araneae, Nesticidae) from the southern Appalachian Mountains

This revision is based on sampling efforts over the past three decades in the southern Appalachian Mountains which have provided Nesticus (Araneae, Nesticidae) collections of approximately 2100 adult specimens from more than 475 unique collecting events. Using a “morphology first” framework we examined recently collected specimens plus museum material to formulate morphology-based species hypotheses for putative new taxa (discovery phase). Using sequence capture of nuclear ultraconserved elements (UCEs) we analyzed 801 nuclear loci to validate new (and prior) morphology-based species hypotheses (validation phase) and reconstructed a robust backbone phylogeny including all described and new species. Sanger sequencing and UCE-bycatch were also used to gather mitochondrial data for more than 240 specimens. Based on our integrative taxonomic framework ten new Nesticus species are herein described, including N. binfordaesp. nov., N. bondisp. nov., N. caneisp. nov., N. cherokeensissp. nov., N. dellingerisp. nov., N. dykemanaesp. nov., N. jemisinaesp. nov., N. lowderisp. nov., N. roanensissp. nov., and N. templetonisp. nov. Previously unknown males are also described for N. bishopi Gertsch, 1984, N. crosbyi Gertsch, 1984, and N. silvanus Gertsch, 1984, as well as the previously unknown female for N. mimus Gertsch, 1984. Based on combined evidence N. cooperi Gertsch, 1984 is placed in synonymy with N. reclusus Gertsch, 1984. Overall, the montane radiation of Appalachian Nesticus reveals a general lack of species sympatry and compelling biogeographic patterns. Several regional Nesticus taxa are rare, microendemic habitat specialists that deserve conservation attention and detailed future monitoring as conservation sentinels.

DNA Barcoding of Lepidoptera Species from the Maltese Islands: New and Additional Records, with an Insight into Endemic Diversity

This work presents the first outcomes resulting from a DNA barcode reference library of lepidopteran species from Malta. The library presented here was constructed from the specimens collected between 2015 and 2019 and covers the genetic barcodes of 146 species (ca. 25% of lepidopterous Maltese fauna), including four newly recorded Lepidoptera species from the Maltese islands: Apatema baixerasi, Bostra dipectinialis, Oiketicoides lutea, and Phereoeca praecox. The DNA reference barcode library constructed during this study was analyzed in conjunction with publicly available DNA barcodes and used to assess the ability of the local DNA barcodes to discriminate species. Results showed that each species occupies a different BOLD BIN; therefore, DNA barcoding was able to discriminate between the studied species. Our data led to the formation of 12 new BOLD BINs—that is, OTUs that were identified during this work—while nearly 46% of the barcodes generated during this study were never recorded on conspecifics, further indicating the uniqueness of genetic diversity on these central Mediterranean islands. The outcomes of this study highlight the integrative taxonomic approach, where molecular taxonomy plays an important role for biodiversity investigation in its entirety.

Phylogenomics of paleoendemic lampshade spiders (Araneae, Hypochilidae, Hypochilus), with the description of a new species from montane California

Hypochilus is a relictual lineage of Nearctic spiders distributed disjunctly across the United States in three montane regions (California, southern Rocky Mountains, southern Appalachia). Phylogenetic resolution of species relationships in Hypochilus has been challenging, and conserved morphology coupled with extreme genetic divergence has led to uncertain species limits in some complexes. Here, Hypochilus interspecies relationships have been reconstructed and cryptic speciation more critically evaluated using a combination of ultraconserved elements, mitochondrial CO1 by-catch, and morphology. Phylogenomic data strongly support the monophyly of regional clades and support a ((California, Appalachia), southern Rocky Mountains) topology. In Appalachia, five species are resolved as four lineages (H.thorelli Marx, 1888 and H.coylei Platnick, 1987 are clearly sister taxa), but the interrelationships of these four lineages remain unresolved. The Appalachian species H.pococki Platnick, 1987 is recovered as monophyletic but is highly genetically structured at the nuclear level. While algorithmic analyses of nuclear data indicate many species (e.g., all H.pococki populations as species), male morphology instead reveals striking stasis. Within the California clade, nuclear and mitochondrial lineages of H.petrunkevitchi Gertsch, 1958 correspond directly to drainage basins of the southern Sierra Nevada, with H.bernardino Catley, 1994 nested within H.petrunkevitchi and sister to the southernmost basin populations. Combining nuclear, mitochondrial, geographical, and morphological evidence a new species from the Tule River and Cedar Creek drainages is described, Hypochilusxomotesp. nov. We also emphasize the conservation issues that face several microendemic, habitat-specialized species in this remarkable genus.