Machine-learning strategies for testing patterns of morphological variation in small samples: sexual dimorphism in gray wolf (Canis lupus) crania

Abstract

Background: Studies of mammalian sexual dimorphism have traditionally involved the measurement of selected dimensions of particular skeletal elements and use of single data-analysis procedures. Consequently, such studies have been limited by a variety of both practical and conceptual constraints. To compare and contrast what might be gained from a more exploratory, multifactorial approach to the quantitative assessment of form-variation, images of a small sample of modern Israeli gray wolf (Canis lupus) crania were analyzed via elliptical Fourier analysis of cranial outlines, a Naïve Bayes machine-learning approach to the analysis of these same outline data, and a deep-learning analysis of whole images in which all aspects of these cranial morphologies were represented. The statistical significance and stability of each discriminant result were tested using bootstrap and jackknife procedures. Results: Our results reveal no evidence for statistically significant sexual size dimorphism, but significant sexmediated shape dimorphism. These are consistent with the findings of prior wolf sexual dimorphism studies and extend these studies by identifying new aspects of dimorphic variation. Additionally, our results suggest that shapebased sexual dimorphism in the C. lupus cranial complex may be more widespread morphologically than had been appreciated by previous researchers. Conclusion: Our results suggest that size and shape dimorphism can be detected in small samples and may be dissociated in mammalian morphologies. This result is particularly noteworthy in that it implies there may be a need to refine allometric hypothesis tests that seek to account for phenotypic sexual dimorphism. The methods we employed in this investigation are fully generalizable and can be applied to a wide range of biological materials and could facilitate the rapid evaluation of a diverse array of morphological/phenomic hypotheses.

What you sample is what you get: ecomorphological variation in Trithemis (Odonata, Libellulidae) dragonfly wings reconsidered

“…From the results we have presented above it is unquestionably clear that the classic GM approach, when applied to the analysis of Trithemis wing-shape data, was the one that performed least well in finding, summarizing, and testing sets of characteristics that could be used to answer the question of whether shape variance was distributed among Trithemis landscape and water-body ecological guilds in a continuous or disjunct manner. What is also clear is that this comparative finding is neither an unusual, nor an exceptional, result (e.g., [ 45 , 46 , 59 , 65 – 74 ]).…”

“…What you sample determines what results you get. Morphometric representations of biological forms, especially those sampled by sparse sets of landmark-semilandmark points, cannot, should not, and must not be mistaken for the morphologies of the individuals or species themselves and the results generated therefrom pertain only to those aspects of the morphology that were sampled, not to the overall morphology itself (see also [ 59 ] where this issue was problematic). This distinction should be kept in mind, especially if negative results are obtained from any morphometric hypothesis test.…”

“…11 ). Recent published applications of this strategy have focused on systems for describing differences between image pairs drawn from large datasets using text-based descriptors [ 90 , 91 ] as well as image-based analyses [ 33 , 45 , 46 , 59 , 71 , 92 ]. In terms of the analysis of small-to-modestly sized samples, there are many advantages to this approach, including relaxation of the use of single assessments of individual forms insofar as all, or most, pairwise comparisons between images in a dataset can be employed for CNN training.…”

Morphology-Based Identification of Bemisia tabaci Cryptic Species Puparia via Embedded Group-Contrast Convolution Neural Network Analysis

“…1 ) precludes the application of more advanced landmark and/or semilandmark-based, morphometric data-collection procedures. Recently, machine-learning (ML) algorithms have been applied to the problem of morphological group-discrimination in which the input data are digital images of the specimens themselves (see MacLeod et al 2005 ; MacLeod 2007 , 2015 , 2018 ; Ranaweera et al 2009 ; Wilf et al 2016 ; Valan et al 2019; Hoyal Cuthill et al 2019 ; MacLeod and Kolska-Horwitz 2020 ). Increasingly, ML procedures have been shown to be capable of delivering morphological group-discrimination results superior to those of even the most advanced GM-style analyses ( MacLeod 2015 , 2018 ; Hoyal Cuthill et al 2019 ; MacLeod and Kolska-Horwitz 2020 ).…”

“…Recently, machine-learning (ML) algorithms have been applied to the problem of morphological group-discrimination in which the input data are digital images of the specimens themselves (see MacLeod et al 2005 ; MacLeod 2007 , 2015 , 2018 ; Ranaweera et al 2009 ; Wilf et al 2016 ; Valan et al 2019; Hoyal Cuthill et al 2019 ; MacLeod and Kolska-Horwitz 2020 ). Increasingly, ML procedures have been shown to be capable of delivering morphological group-discrimination results superior to those of even the most advanced GM-style analyses ( MacLeod 2015 , 2018 ; Hoyal Cuthill et al 2019 ; MacLeod and Kolska-Horwitz 2020 ). Accordingly, the primary aim of this investigation was to determine whether the application of ML and computer vision-based morphological data and data-analysis methods might facilitate the identification of puparium-based morphological distinctions between members of the B. tabaci complex represented by a large set of currently recognized genetic species.…”