Acoustic features as a tool to visualize and explore marine soundscapes: Applications illustrated using marine mammal passive acoustic monitoring datasets.

Passive Acoustic Monitoring (PAM) is emerging as a solution for monitoring species and environmental change over large spatial and temporal scales. However, drawing rigorous conclusions based on acoustic recordings is challenging, as there is no consensus over which approaches are best suited for characterizing marine acoustic environments. Here, we describe the application of multiple machine-learning techniques to the analysis of two PAM datasets. We combine pre-trained acoustic classification models (VGGish, NOAA and Google Humpback Whale Detector), dimensionality reduction (UMAP), and balanced random forest algorithms to demonstrate how machine-learned acoustic features capture different aspects of the marine acoustic environment. The UMAP dimensions derived from VGGish acoustic features exhibited good performance in separating marine mammal vocalizations according to species and locations. RF models trained on the acoustic features performed well for labeled sounds in the 8 kHz range; however, low- and high-frequency sounds could not be classified using this approach. The workflow presented here shows how acoustic feature extraction, visualization, and analysis allow establishing a link between ecologically relevant information and PAM recordings at multiple scales, ranging from large-scale changes in the environment (i.e., changes in wind speed) to the identification of marine mammal species.

To sleep or not to sleep: Dormancy and life history traits in Eucypris virens (Crustacea, Ostracoda).

Dormancy represents an investment with its own costs and benefit. Besides the advantage obtained from the avoidance of harsh environments and from the synchronization of life cycles with seasonal changes, an organism could benefit from a temporary stop in growth and reproduction. To test this hypothesis a transgenerational experiment was carried out comparing the life history traits of clonal females of Eucypris virens from resting and non-resting eggs at two different photoperiods: short day length (6:18 L:D), proxy of favorable but unpredictable late winter-spring hydroperiod, and long day length (16:8 L:D) proxy of dry predictable unfavorable season, inducing resting egg production and within-generation plasticity (WGP). Clonal females that were dormancy deprived showed the highest age at first deposition and the lowest fecundity. Dormancy seems to work as a resetting mechanism of reproduction. Transgenerational plasticity (TGP) had a bounce back pattern: the phenotype of F1 generation was influenced by cues experienced in the F0 generation but the effects of F0 exposure were not evident in the F2. TGP might be adaptive when a mother experiences some kind of seasonality or stochasticity producing both resting and nonresting eggs. A positive relationship between the number of resting eggs and the total number of eggs per females suggested the absence of trade-off between dormancy and reproduction. Both WGP and TGP increase the mother long term fitness with important consequences on population dynamics, on the way a species spread throughout space and time and might respond to climate change.

Molecular Barcoding: A Tool to Guarantee Correct Seafood Labelling and Quality and Preserve the Conservation of Endangered Species.

The recent increase in international fish trade leads to the need for improving the traceability of fishery products. In relation to this, consistent monitoring of the production chain focusing on technological developments, handling, processing and distribution via global networks is necessary. Molecular barcoding has therefore been suggested as the gold standard in seafood species traceability and labelling. This review describes the DNA barcoding methodology for preventing food fraud and adulteration in fish. In particular, attention has been focused on the application of molecular techniques to determine the identity and authenticity of fish products, to discriminate the presence of different species in processed seafood and to characterize raw materials undergoing food industry processes. In this regard, we herein present a large number of studies performed in different countries, showing the most reliable DNA barcodes for species identification based on both mitochondrial (COI, cytb, 16S rDNA and 12S rDNA) and nuclear genes. Results are discussed considering the advantages and disadvantages of the different techniques in relation to different scientific issues. Special regard has been dedicated to a dual approach referring to both the consumer's health and the conservation of threatened species, with a special focus on the feasibility of the different genetic and genomic approaches in relation to both scientific objectives and permissible costs to obtain reliable traceability.

Unsupervised machine learning and geometric morphometrics as tools for the identification of inter and intraspecific variations in the Anopheles Maculipennis complex.

Geometric morphometric analysis was combined with two different unsupervised machine learning algorithms, UMAP and HDBSCAN, to visualize morphological differences in wing shape among and within four Anopheles sibling species (An. atroparvus, An. melanoon, An. maculipennis s.s. and An. daciae sp. inq.) of the Maculipennis complex in Northern Italy. Specifically, we evaluated: (1) wing shape variation among and within species; (2) the consistencies between groups of An. maculipennis s.s. and An. daciae sp. inq. identified based on COI sequences and wing shape variability; and (3) the spatial and temporal distribution of different morphotypes. UMAP detected at least 13 main patterns of variation in wing shape among the four analyzed species and mapped intraspecific morphological variations. The relationship between the most abundant COI haplotypes of An. daciae sp. inq. and shape ordination/variation was not significant. However, morphological variation within haplotypes was reported. HDBSCAN also recognized different clusters of morphotypes within An. daciae sp. inq. (12) and An. maculipennis s.s. (4). All morphotypes shared a similar pattern of variation in the subcostal vein, in the anal vein and in the radio-medial cross-vein of the wing. On the contrary, the marginal part of the wings remained unchanged in all clusters of both species. Any spatial-temporal significant difference was observed in the frequency of the identified morphotypes.  Our study demonstrated that machine learning algorithms are a useful tool combined with geometric morphometrics and suggest to deepen the analysis of inter and intra specific shape variability to evaluate evolutionary constrains related to wing functionality.

Geometric morphometrics and machine learning as tools for the identification of sibling mosquito species of the Maculipennis complex (Anopheles).

Geometric morphometrics allows researchers to use the specific software to quantify and to visualize morphological differences between taxa from insect wings. Our objective was to assess wing geometry to distinguish four Anopheles sibling species of the Maculipennis complex, An. maculipennis s. s., An. daciae sp. inq., An. atroparvus and An. melanoon, found in Northern Italy. We combined the geometric morphometric approach with different machine learning alghorithms: support vector machine (SVM), random forest (RF), artificial neural network (ANN) and an ensemble model (EN). Centroid size was smaller in An. atroparvus than in An. maculipennis s. s. and An. daciae sp. inq. Principal component analysis (PCA) explained only 33% of the total variance and appeared not very useful to discriminate among species, and in particular between An. maculipennis s. s. and An. daciae sp. inq. The performance of four different machine learning alghorithms using procrustes coordinates of wing shape as predictors was evaluated. All models showed ROC-AUC and PRC-AUC values that were higher than the random classifier but the SVM algorithm maximized the most metrics on the test set. The SVM algorithm with radial basis function allowed the correct classification of 83% of An. maculipennis s. s. and 79% of An. daciae sp. inq. ROC-AUC analysis showed that three landmarks, 11, 16 and 15, were the most important procrustes coordinates in mean wing shape comparison between An. maculipennis s. s. and An. daciae sp. inq. The pattern in the three-dimensional space of the most important procrustes coordinates showed a clearer differentiation between the two species than the PCA. Our study demonstrated that machine learning algorithms could be a useful tool combined with the wing geometric morphometric approach.