Search for: All records

Coordinated spawning of marine animals releases millions of planktonic eggs into the environment, known as egg boons. Eggs are rich in essential fatty acids and may be an important lipid subsidy to egg consumers. Our aim was to validate the application of fatty acid and stable isotope tracers of egg consumption to potential egg consumers and to confirm egg consumption by the selected species. We conducted feeding experiments with ctenophores, crustaceans and fishes. We fed these animals a common diet of Artemia or a commercial feed (Otohime) and simulated egg boons for half of them by intermittently supplementing the common diet with red drum (Sciaenops ocellatus) eggs for 10–94 days. Controls did not receive eggs. Fatty acid profiles of consumers fed eggs were significantly different from those of controls 24 h after the last egg-feeding event. Consumers took on fatty acid characteristics of eggs. In fishes and ctenophores, fatty acid markers of egg consumption did not persist 2–5 days after the last egg-feeding event, but markers of egg consumption persisted in crustaceans for at least 5–10 days. Additionally, consumption of eggs, which had high values of δ15N, led to δ15N enrichment in crustaceans and a fish. We conclude that fatty acids and nitrogen stable isotope can be used as biomarkers of recent egg consumption in marine animals, validating their use for assessing exploitation of egg boons in nature.

 

Pollen identification is necessary for several subfields of geology, ecology, and evolutionary biology. However, the existing methods for pollen identification are laborious, time-consuming, and require highly skilled scientists. Therefore, there is a pressing need for an automated and accurate system for pollen identification, which can be beneficial for both basic research and applied issues such as identifying airborne allergens. In this study, we propose a deep learning (DL) approach to classify pollen grains in the Great Basin Desert, Nevada, USA. Our dataset consisted of 10,000 images of 40 pollen species. To mitigate the limitations imposed by the small volume of our training dataset, we conducted an in-depth comparative analysis of numerous pre-trained Convolutional Neural Network (CNN) architectures utilizing transfer learning methodologies. Simultaneously, we developed and incorporated an innovative CNN model, serving to augment our exploration and optimization of data modeling strategies. We applied different architectures of well-known pre-trained deep CNN models, including AlexNet, VGG-16, MobileNet-V2, ResNet (18, 34, and 50, 101), ResNeSt (50, 101), SE-ResNeXt, and Vision Transformer (ViT), to uncover the most promising modeling approach for the classification of pollen grains in the Great Basin. To evaluate the performance of the pre-trained deep CNN models, we measured accuracy, precision, F1-Score, and recall. Our results showed that the ResNeSt-110 model achieved the best performance, with an accuracy of 97.24%, precision of 97.89%, F1-Score of 96.86%, and recall of 97.13%. Our results also revealed that transfer learning models can deliver better and faster image classification results compared to traditional CNN models built from scratch. The proposed method can potentially benefit various fields that rely on efficient pollen identification. This study demonstrates that DL approaches can improve the accuracy and efficiency of pollen identification, and it provides a foundation for further research in the field.

 

Insect herbivory can be an important selective pressure and contribute substantially to local plant richness. As herbivory is the result of numerous ecological and evolutionary processes, such as complex insect population dynamics and evolution of plant antiherbivore defenses, it has been difficult to predict variation in herbivory across meaningful spatial scales. In the present work, we characterize patterns of herbivory on plants in a species‐rich and abundant tropical genus (Piper) across forests spanning 44° of latitude in the Neotropics. We modeled the effects of geography, climate, resource availability, andPiperspecies richness on the median, dispersion, and skew of generalist and specialist herbivory. By examining these multiple components of the distribution of herbivory, we were able to determine factors that increase biologically meaningful herbivory at the upper ends of the distribution (indicated by skew and dispersion). We observed a roughly twofold increase in median herbivory in humid relative to seasonal forests, which aligns with the hypothesis that precipitation seasonality plays a critical role in shaping interaction diversity within tropical ecosystems. Site level variables such as latitude, seasonality, and maximumPiperrichness explained the positive skew in herbivory at the local scale (plot level) better for assemblages ofPipercongeners than for a single species. Predictors that varied between local communities, such as resource availability and diversity, best explained the distribution of herbivory within sites, dampening broad patterns across latitude and climate and demonstrating why generalizations about gradients in herbivory have been elusive. The estimated population means, dispersion, and skew of herbivory responded differently to abiotic and biotic factors, illustrating the need for careful studies to explore distributions of herbivory and their effects on forest diversity.

 

Phytochemical diversity is an effective plant defensive attribute, but much more research has focused on genetic and environmental controls of specific defensive compounds than phytochemical diversityper se. Documenting plasticity in phytochemical richness and plant chemical composition as opposed to individual compounds is important for understanding plant defense. This study outlines a multi-site transplant experiment in Cerrado gallery forests in central Brazil, utilizingPiper arboreum(Piperaceae), a prevalent and widespread neotropical shrub. Clones from four distinct populations were planted either at their origin site or in a different forest. Secondary metabolite composition varied between populations initially and then changed after transplanting. Interestingly, clones with chemical profiles that were distinct from the populations where they were introduced experienced reduced specialist chrysomelid herbivory compared to clones that were more chemically similar to the existingP. arboreumpopulations where they were planted. Specialist Lepidoptera herbivory also declined in clones transplanted to a new forest, but this change could not be ascribed to chemical profiles. In contrast, generalist herbivory was unaffected by chemical dissimilarity and transplanting. This research adds to the expanding body of evidence suggesting that phytochemical diversity is a dynamic trait exerting unique effects on different herbivore guilds.

 

A new species of the rarely collected neotropical microgastrine braconid wasp genusLarissimusNixon, represented previously by only a single described species,L. cassanderNixon, was recovered by the Caterpillars and Parasitoids of the Eastern Andes in Ecuador inventory project.Larissimus nigricanssp. nov.was reared from an unidentified species of arctiine Erebidae feeding on the common bamboo speciesChusquea scandensKunth at the Yanayacu Biological Station near Cosanga, Napo Province, Ecuador. The new species is described and diagnosed fromL. cassanderusing both morphological and DNA barcode data.

 

Identifying the mechanisms by which bacterial pathogens kill host cells is fundamental to understanding how to control and prevent human and animal disease. In the case of Bacillus thuringiensis (Bt), such knowledge is critical to using the bacterium to kill insect vectors that transmit human and animal disease. For the Cry4B toxin produced by Bt, its capacity to kill Anopheles gambiae, the primary mosquito vector of malaria, is the consequence of a variety of signaling activities. We show here that Cry4B, acting as first messenger, binds specifically to the bitopic cadherin BT-R₃G-protein-coupled receptor (GPCR) localized in the midgut of A. gambiae, activating the downstream second messenger cyclic adenosine monophosphate (cAMP). The direct result of the Cry4B–BT-R₃binding is the release of α_sfrom the heterotrimeric αβγ-G-protein complex and its activation of adenylyl cyclase (AC). The upshot is an increased level of cAMP, which activates protein kinase A (PKA). The functional impact of cAMP–PKA signaling is the stimulation of Na⁺/K⁺-ATPase (NKA) which serves as an Na⁺/K⁺pump to maintain proper gradients of extracellular Na⁺and intracellular K⁺. Increased level of cAMP amplifies NKA and upsets normal ion concentration gradients. NKA, as a scaffolding protein, accelerates the first messenger signal to the nucleus, generating additional BT-R₃molecules and promoting their exocytotic trafficking to the cell membrane. Accumulation of BT-R₃on the cell surface facilitates recruitment of additional toxin molecules which, in turn, amplify the original signal in a cascade-like manner. This report provides the first evidence of a bacterial toxin using NKA via AC/PKA signaling to execute cell death.

 

Prediction algorithms, such as deep neural networks (DNNs), are used in many domain sciences to directly estimate internal parameters of interest in simulator-based models, especially in settings where the observations include images or complex high-dimensional data. In parallel, modern neural density estimators, such as normalizing flows, are becoming increasingly popular for uncertainty quantification, especially when both parameters and observations are high-dimensional. However, parameter inference is an inverse problem and not a prediction task; thus, an open challenge is to construct conditionally valid and precise confidence regions, with a guaranteed probability of covering the true parameters of the data-generating process, no matter what the (unknown) parameter values are, and without relying on large-sample theory. Many simulator-based inference (SBI) methods are indeed known to produce biased or overly con- fident parameter regions, yielding misleading uncertainty estimates. This paper presents WALDO, a novel method to construct confidence regions with finite-sample conditional validity by leveraging prediction algorithms or posterior estimators that are currently widely adopted in SBI. WALDO reframes the well-known Wald test statistic, and uses a computationally efficient regression-based machinery for classical Neyman inversion of hypothesis tests. We apply our method to a recent high-energy physics problem, where prediction with DNNs has previously led to estimates with prediction bias. We also illustrate how our approach can correct overly confident posterior regions computed with normalizing flows.