Palatal morphology predicts the paleobiology of early salamanders

Jia, Jia; Li, Guangzhao; Gao, Ke‐Qin

doi:10.7554/elife.76864

Cited by 4 publications

(4 citation statements)

References 77 publications

(127 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) (Schutz et al, 2022), Gasterosteidae (2) (Fraser & El-Sabaawi, 2022), Hominidae (Krenn et al, 2022), Hynobiidae (Jia et al, 2022), Muridae (a/b) (Devine et al, 2022), Poeciliidae (Riesch et al, 2016), Serranidae (Alós et al, 2014), Cichlidae (Ronco et al, 2020), Colubridae+ (Head & Polly, 2015), Crocodylidae (Watanabe & Slice, 2014), Formicidae (Kennedy et al, 2014), Ocypodidae (Hopkins et al, 2016), Percidae (Martin & Mendelson, 2014), Vespidae (Perrard et al, 2014), andViviparidae (Van Bocxlaer &Hunt, 2013). Dataset details are listed in Table S1.…”

Section: Conflict Of Interest Statementmentioning

confidence: 99%

Classifying high-dimensional phenotypes with ensemble learning

Devine

Kurki

Epp

et al. 2023

Preprint

View full text Add to dashboard Cite

Classification is a fundamental task in biology used to assign members to a class. While linear discriminant functions have long been effective, advances in phenotypic data collection are yielding increasingly high-dimensional datasets with more classes, unequal class covariances, and non-linear distributions. Numerous studies have deployed machine learning techniques to classify such distributions, but they are often restricted to a particular organism, a limited set of algorithms, and/or a specific classification task. In addition, the utility of ensemble learning or the strategic combination of models has not been fully explored.We performed a meta-analysis of 33 algorithms across 20 datasets containing over 20,000 high-dimensional shape phenotypes using an ensemble learning framework. Both binary (e.g., sex, environment) and multi-class (e.g., species, genotype, population) classification tasks were considered. The ensemble workflow contains functions for preprocessing, training individual learners and ensembles, and model evaluation. We evaluated algorithm performance within and among datasets. Furthermore, we quantified the extent to which various dataset and phenotypic properties impact performance.We found that discriminant analysis variants and neural networks were the most accurate base learners on average. However, their performance varied substantially between datasets. Ensemble models achieved the highest performance on average, both within and among datasets, increasing average accuracy by up to 3% over the top base learner. Higher class R2values, mean class shape distances, and between– vs. within-class variances were positively associated with performance, whereas higher class covariance distances were negatively associated. Class balance and total sample size were not predictive.Learning-based classification is a complex task driven by many hyperparameters. We demonstrate that selecting and optimizing an algorithm based on the results of another study is a flawed strategy. Ensemble models instead offer a flexible approach that is data agnostic and exceptionally accurate. By assessing the impact of various dataset and phenotypic properties on classification performance, we also offer potential explanations for variation in performance. Researchers interested in maximizing performance stand to benefit from the simplicity and effectiveness of our approach made accessible via the R packagepheble.

show abstract

Section: Conflict Of Interest Statementmentioning

confidence: 99%

Classifying high-dimensional phenotypes with ensemble learning

Devine

Kurki

Epp

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We characterized vomer morphology in the lungless salamander tribe Spelerpini to determine if morphology is taxonomically conserved or driven by ecology, as in cryptobranchoid salamanders (Jia et al, 2022). This clade of lungless salamanders includes both terrestrial and secondarily-aquatic species, presenting an opportunity to consider convergence.…”

Section: Stereochilus Marginatus Semiaquatic Semiaquaticmentioning

confidence: 99%

“…This suggestion was supported by a geometric morphometrics analysis by Darcy (2015). However, a study by Jia et al (2022) that included both terrestrial and aquatic species among cryptobranchid and hynobiid salamanders (Cryptobrancoidea) determined that morphology is constrained by habitat‐determined feeding mechanics (i.e., feeding in terrestrial, semiaquatic, or aquatic habitats) among living and extinct salamander species.…”

Section: Introductionmentioning

confidence: 99%

Morphological variation in the vomer of aquatic and terrestrial spelerpini salamanders

Darcy

Anderson

2023

Journal of Morphology

View full text Add to dashboard Cite

The vomer is an important tooth‐bearing cranial bone in the lungless salamanders (Caudata: Plethodontidae) that serves different functional roles in aquatic versus terrestrial feeding. Vomerine tooth rows that run parallel with the maxillary teeth are thought to help grasp prey while expelling water from the mouth, while posterior extensions of the tooth row may help terrestrial taxa bring prey down the throat. We hypothesize that these two general morphological types will correlate with the habitat (aquatic vs. terrestrial) of adult salamanders. Alternatively, variation in form may be due to taxonomic effects, such that closely related species will have similar vomer morphology regardless of adult habitat. To test this hypothesis, we examined vomer shape on a set of species of the morphologically diverse tribe Spelerpini, in which two of the five genera (Eurycea and Gyrinophilus) include both aquatic and terrestrial species. Data were collected using micro computed tomography (micro‐CT) scans from specimens from the Field Museum of Natural History and the Illinois Natural History Survey; additional data was obtained from public online repositories including Morphosource.org. Two‐dimensional geometric morphometric analyses were performed to capture shape variation of both the vomer and the vomerine tooth row. We found clear separation between aquatic and terrestrial taxa, with most of the variation due to differences in the vomerine tooth row. Differences ascribed to habitat use likely correspond to feeding behavior, and the functional role of the vomer in prey processing warrants further investigation in this species‐rich salamander family.

show abstract

“…There are still many issues related to early salamanders' origins and phylogenetic relationships, partly because only a few fossil species and different characteristics have been selected for further analyses. If poorly characterized taxa are excluded from evolutionary investigations or new fossils with more anatomical features are added, the results of phylogenetic analyses would change drastically, [13][14][15] emphasizing the need for more detailed studies.…”

Section: Introductionmentioning

confidence: 99%

Micro-XRF Mapping Study On The Taphonomy Of A Jurassic Larval Salamander Fossil From Inner Mongolia Of China

Li¹

2022

At.Spectrosc.

View full text Add to dashboard Cite

Salamanders (Urodela) are members of a clade of extant amphibians whose fossil records extend far into the Mesozoic Era and have a classic body model of generalized tetrapods. The characteristic features of larval ontogenetic stages of early fossil salamanders can be used to better understand this group's origins and evolutionary history. This study used non-destructive micro-X-ray fluorescence (micro-XRF) imaging technology to explore the taxonomy and patterns of elemental concentrations of a Middle-Late Jurassic larval salamander fossil collected from northeastern China. P distribution mapping of the skeletal structure indicated that the fossil belongs to Jeholotriton paradoxus. Further analysis revealed that the skeleton is enriched with P, while the soft-tissue remains are enriched with Fe, Mn, Ga, and Cr. Ni and Zn in the trunk area come from the external environment. The Yanliao biota, to which the fossil belongs, contains many salamander fossils. Further XRF research will reveal more information about the ontogenesis and taphonomy of early salamanders.

show abstract

Palatal morphology predicts the paleobiology of early salamanders

Cited by 4 publications

References 77 publications

Classifying high-dimensional phenotypes with ensemble learning

Classifying high-dimensional phenotypes with ensemble learning

Morphological variation in the vomer of aquatic and terrestrial spelerpini salamanders

Micro-XRF Mapping Study On The Taphonomy Of A Jurassic Larval Salamander Fossil From Inner Mongolia Of China

Contact Info

Product

Resources

About