Texas A&M University – Kingsville

Abstract Live capturing of wildlife is important for research and conservation, but is rife with challenges including animal welfare concerns, resource commitments (e.g., time and effort), and low capture success rates. Such challenges may be especially apparent for species listed under the Endangered Species Act and species whose ecology impacts capture success. Thus, it is important to investigate and implement strategies that can improve the capture success of endangered carnivores while reducing human effort and risk to animals. In the United States, ocelots ( Leopardus pardalis ) are a federally endangered species. To minimize risk to ocelots, capture efforts for research in Texas are limited to box trapping, but capture success is generally low. We tested whether visual and scent lures can increase capture success of ocelots and other co‐occurring mesocarnivores with box traps baited with a live bird on a private ranch in the Texas‐Tamaulipan thornscrub ecosystem. From December 2023 through April 2024, we used a randomized design and placed visual (hanging compact disc [CD] or shiny ribbon) and scent (civet musk or ocelot urine) lures at a total of 20 box traps and established 5 control traps across 3 temporally and geographically discrete trapping sessions for 75 trapping locations. We used linear regression to assess 2,900 trap nights and found that none of the lures (scent nor visual) impacted capture success of ocelots, bobcats (Lynx rufus ), or coyotes (Canis latrans ). Further, lures did not increase or decrease capture of non‐target animals. Based on our findings, we cannot recommend the use of CDs, ribbon, civet musk, or ocelot urine as lures to increase ocelot box trapping success, as they did not increase capture success of ocelots or other target carnivores nor decrease captures of non‐target species. We suggest that researchers prioritize other strategies to increase capture success (e.g., setting additional traps) rather than the addition of lures we tested.

Electronic cigarettes (e-cigs) are major contributors to inflammatory-mediated responses, which are implicated in a vast array of pathophysiological conditions, including cardiovascular disease (CVD). More recently e-cigs have been recognized as a source of thirdhand exposure (THEC); the process by which expelled toxins settle on materials (i.e., carpets, curtains, clothes etc.) and undergo chemical reactions, rendering them more harmful overtime. Herein, mice were exposed to THEC for four months, and platelet reactivity, systemic mediated effects on platelet function, and cytokine expression profiles were analyzed in both sexes. Our data revealed a hyperactive platelet phenotype as determined by shortened bleeding and occlusion times, enhanced platelet aggregation, and dense granule secretion with no significant difference between males and females. Cytokines, amongst other inflammatory molecules, are well documented mediators by which platelet function is modulated and they also enhance susceptibility to CVD. To this end, and to elucidate the mechanism by which platelet reactivity was augmented, washed platelets that were exposed only to clean air (CA) and resuspended in THEC exposed plasma, displayed significantly increased platelet aggregation, dense granule secretion, and p-selectin expression. Indeed, this data suggests that THEC exposure elicits a systemic effect, enhancing platelet response, and was further validated by a dysregulated cytokine profile using plasma, free of platelets, in a sex-dependent manner. Collectively and for the first time, we highlight that both males and females are at similar risk of THEC-mediated prothrombotic phenotype, which is underlined-at least in part- by an indirect systemic effect of exposure on platelet reactivity that involves changes in the cytokine profile. These findings underscore this form of exposure as a threat to cardiovascular health.

Abstract Buteo plagiatus (Gray Hawk), state-listed as threatened in Texas, occupies a narrow, increasingly fragmented portion of its range in the United States, where ongoing habitat loss may jeopardize long-term persistence. To quantify resource selection, we used GPS-tracking data and resource selection functions for adults and fledglings in the Lower Rio Grande Valley of Texas, modeling them separately at population and individual levels. We found that across life stages, seasons, and sexes, B. plagiatus consistently selected areas with taller trees, with differences by life stage and scale of selection. Tree height and tree height variability, rather than land cover classes or proximity to riparian zones, were the strongest predictors of use. In contrast with our hypotheses, distance to water, a presumed proxy for riparian zones, was not a consistent predictor of use. Within their home range, both age classes favored taller trees, with adults also responding to tree height variability, but fledglings exhibited greater individual variability. At the nest-site scale, nests were located in wooded areas with greater tree height variability, with taller trees (β = 0.37), less barren land cover within 1,500 m (β = −0.56), and closer to cropland (β = −0.70), all strongly supported predictors of nest-site placement. Many were also near urban areas with higher proportions of developed land (β = 0.31). Individual-level models highlighted variability in selection, especially at fine scales, emphasizing the importance of accounting for behavioral diversity in conservation planning. Our findings suggest that protecting and restoring wooded patches with taller trees and greater tree height variability, including those within urban areas, will be critical for sustaining B. plagiatus at the northeastern edge of its range, where landscapes face continued degradation and fragmentation, and conservation actions must balance pressures from urban expansion, agriculture, and alteration of riparian areas.

Early and reliable identification of intracranial tumors from MRI can shorten time to treatment and reduce invasive diagnostic procedures. This study investigates supervised deep learning for four-way classification of brain MRI into glioma, meningioma, pituitary tumor, and no-tumor using a curated dataset of 7,023 T1-weighted images split into disjoint training and test sets. Five modern convolutional backbones MobileNetV2, EfficientNetB0, ResNet50, DenseNet121, and ConvNeXt-Tiny were fine-tuned with identical preprocessing stratified sampling and class balanced mini batches to ensure fair comparison. To improve generalization the pipeline applied contrast limited adaptive histogram equalization intensity standardization center crop and resize to 224×224 and light spatial photometric augmentation training used label smoothing cosine learning rate decay and early stopping. Among the evaluated models, ConvNeXt-Tiny produced the most consistent performance on the held-out test set, achieving an overall accuracy of 0.96, an F1-score of 0.96. While EfficientNetB0 and DenseNet121 offered competitive accuracy with fewer parameters, ConvNeXt-Tiny demonstrated superior robustness in distinguishing tumor subtypes. MobileNetV2 provided the best compute-latency trade-off for edge deployment. The results indicate that careful normalization and balanced fine-tuning narrow the gap between compact and heavier architectures, and that modern ConvNeXt designs can yield strong accuracy without excessive model size.

The rapid evolution of technology has significantly transformed supply chain management and finance, with blockchain and big data emerging as pivotal drivers of innovation. This paper explores how the integration of blockchain and big data enables predictive analytics and financial metrics to revolutionize global supply chain networks. Blockchain enhances transparency, traceability, and security, while big data analytics provides actionable insights for optimizing supply chain processes and financial decision-making. By leveraging predictive analytics, stakeholders can anticipate demand fluctuations, mitigate risks, and improve operational efficiency. Furthermore, the fusion of financial metrics with advanced analytics facilitates more accurate forecasting, enhanced liquidity management, and streamlined supply chain financing. This study also examines real-world applications and case studies, highlighting the benefits of these technologies in creating resilient, efficient, and financially sustainable supply chain systems. The findings emphasize the potential of blockchain and big data as catalysts for driving innovation and ensuring long-term competitiveness in global markets.