Aims: The present study aims to assess the capabilities, limitations, and practical considerations of integrating Large language models alongside pharmacists. Specifically, the performance of four LLMs was evaluated in responding to related queries for oral Medication dosage prescriptions across different age Groups. Method: The questions were categorized into seven domains. These questions were selected based on the most frequently prescribed drug referenced in global guidelines. Three questions and a case scenario per question were designed for each domain. Prompts were written using the zero-shot method, and collected responses were assessed based on five key factors: Response rate, Accuracy, Completeness, Clarity, and Safety, by comparing them against UpToDate. Results: None of the LLMs had direct access to UpToDate. However, GPT-4o responded correctly to all of the case-based questions. While GPT4o achieved the highest performance, results showed Copilot significantly weaker than it (P<0.05). Meanwhile, the lowest Response rate was observed in Gemini1.5Pro, while Copilot ranked last. Additionally, all LLMs were Safe except Copilot and Claud3.5sonnet V2, which produced unsafe and hazardous responses. Discussions: Findings underscore that while LLMs avoided answering or provided incomplete information, GPT-4o has had promising performance in handling both simple and complex queries in direct and case-based questions. Our results highlight the need to consider specific conditions before the wider integration of LLMs in pharmaceutical practice.

Background and Purpose Adjuvant chemotherapy with 5-fluorouracil (5-FU) prodrugs is now the standard of care for stage I–III oral squamous cell carcinoma (OSCC); however, it remains to be clarified for which patient neoadjuvant chemotherapy with 5-FU prodrugs may be an option. Experimental Approach In the Ryukyu cohort [neoadjuvant chemotherapy with 5-FU prodrugs (UFT or S-1) + bleomycin, neoadjuvant chemotherapy with UFT + bleomycin + cisplatin, and up-front surgery] and Gunma cohort [neoadjuvant chemotherapy with S-1 and up-front surgery], the incidence of distant metastasis was significantly higher in male patients or those with less well differentiated tumors. Key Results Based on these data, we compared the overall survival of male patients who received neoadjuvant chemotherapy and those who received up-front surgery in each cohort. Neoadjuvant chemotherapy with S-1 or S-1 + bleomycin significantly increased overall survival rates in male patients. The ratio of male patients correlated with less well differentiated tumors. Conclusion and Implications Data in The Cancer Genome Atlas revealed that the expression of DNA repair genes correlated with male sex and less well differentiated tumors, while higher Myd88 expression and type I interferon activity were associated with female sex and well-differentiated tumors.

Background: In CKD cases, early detection is disparaging to prevent this disease or related adverse renal conditions. Early identification also lessens related morbidity & mortality. Machine Learning provides effective approaches that helps to intensify the test validity, also enables premature intervention. Objective: This study aims to review and compare existing machine learning models used for chronic kidney disease (CKD) prediction and to experimentally evaluate selected algorithms—Logistic Regression, Support Vector Machine, and Random Forest—on a real-world clinical dataset to validate their performance and interpretability. Methods: In first phase literature review on 20 relevant studies was conducted to extract information like key ml algorithms, performance, and features. In the second phase we conducted an experimental comparison using a public CKD dataset on Kagle. After preprocessing the data, the model of ML like SVM, Logistic regression and Random forest were trained and evaluated by accuracy, precision, recall, F1 score, and ROC-AUC as performance metrics. All the analysis were performed using Python 3.11 in Google Colab environment. Results: Literature review showed ML model accuracy ranging 95 to 100% for CKD prediction. In our Experimental comparison, random Forest outperformed others with perfect scores across all metrics (Accuracy, Precision, Recall, F1 Score, ROC-AUC = 1.00). Logistic Regression and SVM also demonstrated strong results, with F1 scores of 0.9899 and 0.9796, respectively Conclusion: The findings confirm that machine learning models—especially ensemble methods like Random Forest—are highly effective for early CKD prediction. The experimental results align with prior literature and highlight clinically relevant features, demonstrating the potential of ML to support diagnostic decision-making in nephrology.

This study presents a comprehensive analysis of stability, bifurcation, and chaotic dynamics in the generalized Koren-Feingold cloud-rain system, incorporating time delay in rain formation. The model is analyzed analytically by determining the existence and stability of equilibrium points, proving the occurrence of the double Hopf bifurcation. Our results demonstrate that time delay plays a crucial role in dictating the dynamical transitions of the system. A small delay maintains system stability, whereas an increase beyond a critical threshold induces periodic oscillations through a Hopf bifurcation. Further increments lead to period-doubling bifurcations, chaotic attractors, and intermittent oscillatory regimes, demonstrating the formation of intricate meteorological phenomena. Numerical simulations reinforce these findings, illustrating that prolonged delays can lead to persistent cloud formation, irregular precipitation patterns, and chaotic fluctuations in the dynamics of rain. These insights deepen our understanding of cloud-rain interactions and provide valuable implications for improving weather prediction models, particularly in capturing extreme precipitation events influenced by atmospheric memory effects.

Microbiome composition and function are strongly influenced by its environmental factors, with major shifts driven by intensified anthropogenic pressures over the past centuries. This timeframe extends beyond the scope of traditional experimental and longitudinal studies commonly used to investigate microbiome dynamics. The vast collection of historical samples available in museums and herbaria around the world represent a largely untapped resource for exploring host-microbiome interactions across broader temporal and spatial scales. However, their potential remains underutilized due to incompatibilities with standard analytical pipelines and limited knowledge of optimal classification parameters. While host DNA removal has traditionally considered essential for accurate taxonomic assignment of metagenomic reads, this step is often impractical for many historical samples due to the lack of reference genomes for their host species. Here, we demonstrated that host DNA content does not significantly affect key microbial ecological metrics such as alpha- and beta-diversity. Additionally, metagenomic reads from historical samples are often highly fragmented due to post-mortem degradation. Using k-mer analysis of genomic sequences from hosts and their associated microbiomes, we show that reads as short as 21 bp can still produce reliable results, enabling the recovery of microbial signals that would otherwise be discarded. Overall, this study provides a solid foundation for incorporating natural history collections into host-associated microbiome research, offering valuable insights into the long-term effects of anthropogenic change on microbial communities.

Iridium (Ir)-based materials are promising candidates for acidic oxygen evolution reaction (OER) but face challenges such as high cost and aggregation. In this study, we synthesized a low-Ir-content catalyst (1.47 at. %) via electrodeposition, where barium (Ba) doping introduces compressive strain to optimize Ir active sites while mitigating Ir aggregation into clusters or nanoparticles. Structural analyses, including X-ray diffraction (XRD), scanning transmission electron microscopy (STEM), and extended X-ray absorption fine structure (EXAFS), confirm atomic-level dispersion of Ir and Ba, lattice contraction, and shortened Co-O bonds. X-ray photoelectron spectroscopy (XPS) and X-ray absorption near-edge structure (XANES) reveal electron transfer from Ir to Co/Ba, elevating Ir’s oxidation state and enhancing OER activity. In 0.5 M H2SO4, the IrBa-Co3O4 catalyst achieves 10 mA cm-2 at an overpotential of 251 mV and operates stable for 100 hours, outperforming most reported spinel type catalysts. In-situ Raman spectroscopy and XANES attribute the improved kinetics to compressive-strain-induced octahedral Co-oxygen (Cooct-O) bond shortening and optimized Ir-O-Ba/Co coordination. This work demonstrates a strategy for designing cost-effective, durable acidic OER catalysts through synergistic doping and strain engineering.

Title PageTitle: Genetic and Environmental Contributions to Acute (Myo)Pericarditis: A Comprehensive Case StudyAuthors Names: Dr. Destiny Monet Nicholson, PharmD, Lusia Fomuso, PharmD, MPH, CPGxAffiliations: RxKnowledge PLLC, USA

Title PageTitle: Pharmacogenomics in the Hispanic PopulationAuthors Names: Hortencia Monreal, PharmD candidate 2025, Lusia Fomuso, PharmD, MPH, CPGxAffiliations: RxKnowledge PLLC, USA

Studies to describe prokaryotic communities in cycad species have a long history, addressing different methods and techniques to unveil the existing biotic interactions. To date, the interest in describing the microbial community in cycads species continues, mainly focusing on the coralloid root endosphere and rhizosphere microbiomes, particularly emphasizing in its predominant biotic interaction with Cyanobacteria. No studies about the leaf endosphere microbiome composition on cycads has been carried out. In this study, we focused on the microbiome evaluation of six Mexican Dioon species to get insights in the microbiome composition across different biogeographic regions in Mexico, as well as the effect of the environmental variation in the alpha and beta diversities. Overall, we observed that prokaryotic endophytes alpha diversity showed a clear association with genetic and demographic reports for most Dioon species analyzed. Also, a strong association of microbiome diversity patterns and environmental variation experienced by Dioon hosts was found, in line with the historical aridification processes that could drive actual genetic diversity in Dioon species and which possible influence the present-day composition and diversity of cycads leaf endosphere microbiomes.

The rapid advancement of Artificial Intelligence (AI) has opened new avenues for enhancing the capabilities of Information and Communication Technology (ICT) infrastructure. As digital systems become increasingly complex, traditional management approaches often struggle to maintain efficiency, security, and scalability. This study investigates the integration of AI technologies into ICT infrastructure, focusing on how machine learning, intelligent automation, and predictive analytics are transforming key areas such as network optimization, data management, cybersecurity, and system maintenance. Through a combination of literature review and case study analysis, the research identifies the benefits, challenges, and practical implications of AI implementation. Findings suggest that AI can significantly improve decision-making, reduce downtime, and optimize resource allocation within ICT systems. However, successful integration requires addressing issues related to data quality, algorithm transparency, and workforce adaptation. The paper concludes with strategic recommendations for adopting AI in ICT environments to foster resilient, adaptive, and intelligent digital infrastructure.

Tiempo sugerido: 50 minutosDescripción : En esta lección aprenderemos paso paso sobre el tratamiento de aguas residuales y su beneficio para la comunidad.

Natural fibres from renewable resources offer a sustainable and biodegradable alternative to synthetic reinforcements. This study investigates the thermal and mechanical properties of Borassus husk fibre/epoxy composites, fabricated via the hand layup process using 5% NaOH alkali treatment at varying durations (0.5-2 hours). Their thermal and thermo-mechanical properties were investigated through thermogravimetric analysis (TGA) and dynamic mechanical analysis (DMA) followed by outgassing test. Results indicate that alkali treatment significantly enhances the thermal stability of the composites, as evidenced by increased char content (up to 8.11%) and higher integral procedural decomposition temperature (IPDT), with the 0.75-hr treated fibre/epoxy achieving the highest IPDT (525°C). The composites also demonstrated superior energy dissipation and mechanical stiffness compared to neat epoxy (NE) and other bio-fibres based composites. The glass transition temperature (Tg) decreased from 150°C (NE) to 126°C (0.5TBHFE)-137°C (0.75TBHFE), yet outperforming composites reinforced with other conventional natural fibres. Additionally, storage modulus and damping factor (tan δ) improved significantly, with 0.5TBHFE exhibiting the best balance between stiffness and damping. The total mass loss (TML) from outgassing test was increased (0.7-0.89%) considerably compared to NE (0.26%), still confirming acceptable thermal stability. These findings suggest that alkali-treated Borassus husk fibre/epoxy composites offer excellent thermal resistance, mechanical strength and impact resistance, making them promising materials for high-performance applications in the aerospace industries, which would also promote sustainable development. However, variations in properties of biofibres require further research, along with the development of an efficient supply chain for industrial-scale production.

Ear biometrics is a growing field in biometric recognition, offering a unique and stable alternative for personal identification. Unlike traditional biometric modalities such as fingerprints and facial recognition, ears provide a non-intrusive, stable, and unique feature set that remains relatively unchanged over time. Recent advancements in machine learning and deep learning have significantly improved the accuracy of ear recognition systems. This paper explores various methodologies for feature extraction, including geometric-based, texture-based, and deep learning approaches. Additionally, we discuss recognition algorithms such as Principal Component Analysis (PCA), Support Vector Machines (SVM), and Convolutional Neural Networks (CNNs). The applications of ear biometrics extend to security, forensic science, and healthcare, offering promising potential in identity verification and crime investigation. However, challenges such as dataset limitations, occlusions, and variations due to aging persist. Future research should focus on improving deep learning models, increasing dataset diversity, and exploring multi-modal biometric systems for enhanced security and accuracy. This paper provides an in-depth analysis of the advancements, challenges, and future directions in ear biometrics, paving the way for further innovation and real-world applications.

This short communication develops a variational formulation for modeling a silicon semiconductor design through a multi-well approach utilizing phosphorus atoms as a dopant substance. The results are based on standard tools of calculus of variations and optimization theory.

IntroductionRoutine monitoring of biological communities is integral to characterizing ecosystem health, biodiversity, and providing information necessary for public health and resource management (Canonico et al., 2019; Forio and Goethals, 2020). Traditionally, researchers have relied on labor-intensive and invasive techniques, such as netting, trapping, and visual identifications, to identify and quantify species presence. While effective, these conventional methods often present limitations, including constrained spatial coverage, inadequate temporal resolution, disturbance to sensitive habitats, and inability to capture the breadth of organisms present in the environment. In addition, the expertise and methodology required to conduct surveys is highly dependent on the organisms in question. In recent years, environmental DNA (eDNA) analysis has emerged as a complementary approach to traditional observational techniques (Rourke et al., 2022; Stat et al., 2017; Wang et al., 2024; Westgaard et al., 2024).eDNA analysis involves the collection and analysis of genetic material shed by organisms into their surrounding environment (Taberlet et al., 2018). Multicellular organisms release DNA and RNA in a myriad of ways, such as by shedding skin cells, scales, mucus, feces, and gametes, all of which can be extracted and sequenced to identify species present within a given ecosystem. This approach offers several advantages over traditional survey techniques, including non-invasiveness, high sensitivity, and capacity to detect rare or elusive species—particularly those that are difficult to observe visually (Gold et al., 2021; Holman et al., 2019; Noble-James et al., 2023). Moreover, eDNA analysis enables comprehensive assessments of biodiversity over large spatial and temporal scales, providing valuable insights into community composition, species richness, and ecological dynamics (Preston et al., 2024; Searcy et al., 2022; Thomsen et al., 2012).One of the main benefits of aquatic-based omics’ research, responsible for its expanding uptake, is the simplicity in collecting samples from diverse habitats. However, this is also a pivotal challenge to collect samples over relevant temporal and spatial scales. Traditional biological observations are labor and resource-intensive, and while analysis of eDNA data can be complex, eDNA filtration and preservation is universally-accessible. Coverage using conventional eDNA sampling techniques can be comprehensive in marine settings (e.g., coastal) accessible by humans, but laborious as it typically involves sampling in the field with time consuming filtrations in the lab. In contrast, sampling in remote environments requiring crewed ships, e.g., remains limited in spatial and temporal resolution, hindering the characterization of aquatic ecosystems. To address these challenges and leverage the ease of eDNA sample collection, autonomous samplers andin situ biomolecular sensors have emerged, offering a paradigm shift in understanding ecosystem dynamics broadly(McQuillan and Robidart, 2017)(McQuillan and Robidart, 2017; Woods Hole Oceanographic Institution, 2023). These new technologies not only enhance the precision and frequency of data collection but also enable researchers to delve deeper into the molecular mechanisms governing aquatic life in remote locations.A critical but often overlooked consideration in the development of these technologies is the alignment between the diversity of end-user needs and the diversity of sampling systems. No single technology can meet the demands of all applications, and design choices often reflect trade-offs between performance, usability, and cost. For instance, real-time detection capabilities are particularly valuable for event-based sampling, such as during harmful algal blooms or pathogen outbreaks, whereas they are less critical for long-term biodiversity surveys. High-capacity, autonomous sampling systems are better suited for extended deployments or high-frequency data collection, but may be excessive for short-term missions such as ROV-based exploration. Likewise, affordability is a key driver for community science and resource-limited monitoring programs, while fit-for-purpose tools may be prioritized by structured research initiatives, including long-term ecosystem observatories or oceanographic expeditions. Recognizing this diversity of applications, sampling environments and technology helps clarify why no single approach is universally optimal, and highlights the importance of a flexible and interoperable instrumentation landscape.This review summarizes information shared during the Marine ‘Omics Technology and Instrumentation Workshop which was held October 10-12, 2023, supplemented with a subsequent literature review to synthesize the state of autonomous eDNA sampling technology and instrumentation. We explore the latest advancements in autonomous sampling instrumentation, including their design and capabilities, but limit this review to automated samplers, without consideration of the parallel expansion of passive eDNA sampling technologies (Bessey et al., 2021). Additionally, we discuss the integration of these sampling devices with various platforms, advanced in situ analytical capabilities, environmental sensors, and imaging technologies, that collectively enhance the effectiveness and utility of the sampling systems. Finally, we examine current challenges and opportunities associated with autonomous eDNA sampling, including applications, validation, and standardization, all of which are required for a coordinated and larger scale adoption of eDNA (Agersnap et al., 2022; Kelly et al., 2024).

This paper focuses on the gasification dissolution loss experiment of coke by H2O. The results indicate that: As the H2O content increases, the coke surface becomes noticeably rougher, the changes on coke surface can be divided into two situations. The peaks on coke surface are consumed again and again, the coke surface is gradually eroded in this cycle. The degree of unevenness on coke surface shows that the baseline of gasification reaction has been decreased as the reaction progresses. After 60 minutes of gasification, the overall average of pores has expanded, and the distribution of pores on coke surface is no longer uniform. The proportion of isotropic structure decreases with the prolongation of gasification dissolution loss time and temperature. Even in anisotropic structure, the proportions of each structure are changing, and this evolution may be the fundamental reason for the changes in macroscopic properties of coke. In the infrared spectral curve, aromatic ring (C=C stretching vibration) and free hydroxyl group (stretching vibration of O-H) have been discovered. The number of free hydroxyl group has decreased due to the consumption of free hydroxyl group in the gasification reaction. High temperature promotes the separation of free hydroxyl group from H2O, There are more media provided, which is the microscopic reason for the intense gasification. The entire process of coke dissolution by H2O is summarized, the gasification dissolution mechanism based on active sites is proposed to understand the behavior of coke. On active sites, the adsorption of reactants first occurs.

Atrial fibrillation (AF) is a prevalent arrhythmia linked to increased risks of stroke, heart failure, and mortality. Conventional fixed-dose beta-blocker strategies for AF rate control, such as atenolol, fail to account for dynamic individual variability in heart rate (HR) and response. This paper introduces a conceptual framework for a wearable-based precision management system in AF, inspired by continuous glucose monitoring and precision hypertension models. The proposed system utilizes a wrist-worn HR monitor paired with a smartphone app to inform adaptive atenolol dosing in real-time. By employing machine learning for AF detection and adaptive algorithms to modulate dosing based on HR patterns, this approach aims to optimize rate control while minimizing side effects. A pilot study design

Background and objectives: Childhood maltreatment is strongly associated with major depressive disorder (MDD) in adulthood, social support may mediate the effect of childhood maltreatment on MDD, but there were few studies exploring the mediating role of different types of social support between childhood maltreatment and MDD, and whether sex difference existing in the mediating models was also unclear. Methods: The study included 965 MDD patients and 443 healthy participants from nine centers in China. MDD patients are diagnosed using the DSM-IV. The Childhood Trauma Questionnaire-Short Form, Social Support Rate Scale, and 17-item Hamilton Depression Rating Scale, were used to assess the status of childhood maltreatment, social support and MDD, respectively. Multivariable mediation analyses were used to examine the mediating role of different types of social support between childhood maltreatment and MDD after controlling for sex, age, ethnicity, educational level, family history, work status and marriage. Results: The childhood maltreatment increased the risk of MDD. All types of social support significantly reduced the risk of MDD. Total social support (β = -0.37, 95%CI: -0.43, -0.30, p <0.001), perceived support (β = 0.05, 95%CI: 0.03, 0.07) and support utilization (β = 0.01, 95%CI: 0.01, 0.12) played a mediating role in childhood maltreatment and MDD, respectively,but not tangible support. For males, only perceived support mediated the relationship between childhood maltreatment and MDD, while for females, both perceived support and support utilization mediated the relationship.

Objective: We tried to examine the relationships between within-strategy emotion regulation tactics and symptoms of depression and anxiety in adolescents by conducting five meta-analyses. The examined tactics were emotional and situational acceptance, behavioral and cognitive distraction, non-judging and non-reactive mindfulness, behavioral and cognitive problem solving, and positive and relativizing reappraisal. The main question of our endeavor was whether the magnitudes of effect sizes differ between sibling tactics albeit being subtypes of the same emotion regulation strategy. Method: Scientific databases were searched to identify articles with adolescent samples (mean age 10 to 19 years; age range 8 to 21) in which bivariate correlations for at least one relationship of interest were reported. Results: We analyzed a total of 91 articles (92 studies) with 343 effect sizes and found differences in effect size magnitudes across all tactic siblings. The effect sizes ranged from -.46 to .01. Post-hoc analyses revealed that many effect size differences between sibling tactics depend on the particular emotion regulation questionnaire involved. Conclusion: Overall, the results imply the potential clinical importance for a context-dependent selection of emotion regulation tactics above and beyond strategies, and the need for a general tactic flexibility.

This paper presents a new single-switch flyback converter with passive leakage recovery snubber comprising of three diodes, a clamp capacitor, and a small transformer, rated only 5% of the main transformer. The passive snubber enables recovery and direct transfer of main transformer leakage energy to output and soft-transition of active switch over wide load range from full load to 20% load. The power diodes turn on and commutate softly. Complete elimination of semiconductor switching loss enabled high frequency and high efficiency converter operation. Steady state operation, brief design guidelines and detailed performance analysis in terms device stress, soft-switching range, loss estimation, and comparison analysis of proposed topology with recently published other converters are presented in this work. Finally, experimental validation is done through laboratory testing on a 22V-26V input, 72V/120W output prototype converter operating at 100 kHz switching frequency. The experimental observations clearly endorsed the theoretical predictions.

Multiple seas in East Asia have played distinct roles during the Quaternary climatic cycles, which have repeatedly isolated and reconnected temperate forest species, while it remains unclear whether their roles differ. In this study, we used Smilax sieboldii, a widely distributed species along the eastern coast of East Asia, to simultaneously evaluate the roles of multiple seas, including the East China Sea, the Yellow-Bohai Sea, the Korea-Tsushima Strait, and the Taiwan Strait, as geographic barriers or dispersal corridors during historical sea-level fluctuations. We employed Bayesian clustering analysis and demographic simulations to elucidate the genetic structure and evolutionary history. The effects of spatial or environmental differences on population structure were examined through isolation-by-distance (IBD) and isolation-by-environment (IBE) tests. Further, genetic differentiation and gene flow were used as indicators to assess the roles of different seas as barriers or corridors. A pronounced phylogeographic structure was observed in S. sieboldii, with populations divided into three distinct gene pools separated by the East China Sea and the Korea-Tsushima Strait, accompanied by significant genetic admixture at the lineage boundaries. The lineage divergence occurred during the early Quaternary, while secondary contact began in the most recent interglacial period. During population differentiation, the East China Sea and the Korea-Tsushima Strait acted as effective geographic barriers, whereas the Taiwan Strait and the Yellow-Bohai Sea functioned more as dispersal corridors and facilitated greater gene flow. Meanwhile, IBD rather than IBE explained the population structure of S. sieboldii. To conclude, the phylogeographic patterns of S. sieboldii resulted from population isolation and admixture due to sea-level fluctuations since the Pleistocene, and the spatial scale of a sea largely determined its ecological role among the multi-sea systems. These findings improved our understanding of how paleoclimate changes and geological transformations have shaped the speciation and diversification of temperate forest species in East Asia.

Wildlife disease surveillance helps in protecting public health, agriculture, and biodiversity. Planning effective surveillance involves strategic methods for identifying an effective sampling design for a program’s objectives. Gaps in existing standards and complexity for wildlife surveillance justify a need for tools that can build statistically-based intuition in wildlife professionals when designing surveillance systems. To address this need, we present the use of plug-and-play tools, specifically our Surveillance Analysis and Sample Size Explorer (SASSE), to allow wildlife professionals to build intuition about the role of sample size vis-a-vis sampling design and diagnostic test performance in wildlife systems. SASSE uses open-source software (R, R Shiny) to design an interactive, module-based teaching tool to cover key surveillance objectives including detection, prevalence, and epidemiological dynamics. Our tool fills the following gaps: 1) allows a broad audience to apply statistical sample size theory for designing disease surveillance, and 2) provides a simple statistical tool for addressing challenges with disease surveillance design in wildlife populations. Thus, the tool we present here can be used readily to identify efficient sampling designs for a surveillance objective across a wide variety of settings.

Energy extraction and development are fragmenting the landscape in Canada’s oil sands region, creating patches of boreal forest connected by millions of kilometres of cleared linear features. The impacts of oil and gas disturbance on some wildlife species, like caribou and wolves, have been a topic of much research; yet, the influence of energy development on other species, like coyotes – which have recently expanded into the boreal forest and established strong populations – are not well understood. Here, we assessed the effects of linear features on coyote distribution and interspecific interactions, by deploying camera traps across multiple landscapes of varying energy disturbance intensities. Using an information theoretic approach, we competed hypotheses about the effects of linear feature type and density, natural feature coverage, and prey and competitor relative abundances on coyote monthly occurrence. High densities of wide linear features, and high relative abundances of small mammal prey and large competitors, best-predicted coyote occurrence, while natural features had a negative effect. Selection for higher densities of these features suggest that wide linear clearings, like roads and geo-survey seismic lines, provide movement paths for coyotes as they do for wolves, although they may also provide prey subsidies. Snowshoe hare and red squirrel prey, but not ungulates, had a strong positive effect on coyote occurrence, although coyote-prey relationships could shift with the hare cycle. Coyotes appeared to coexist with wolf and lynx competitors, perhaps through shared use of abundant resources and temporal segregation or mediated by large coyote populations – potentially indicating a departure from top-down coyote suppression by dominant heterospecifics. Energy development has fundamentally reshaped the boreal forest of Canada’s oil sands region, giving way to landscapes that support generalist, range-expanding species like coyotes, and altering community dynamics.

This study focuses on transforming real-world scenery into Chinese landscape painting masterpieces through style transfer. Traditional methods using convolutional neural networks (CNNs) and generative adversarial networks (GANs) often yield inconsistent patterns and artifacts. The rise of diffusion models (DMs) presents new opportunities for realistic image generation, but their inherent noise characteristics make it challenging to synthesize pure white or black images. Consequently, existing DM-based methods struggle to capture the unique style and color information of Chinese landscape paintings. To overcome these limitations, we propose CLPFusion, a novel framework that leverages pre-trained diffusion models for artistic style transfer. A key innovation is the Bidirectional State Space Models-CrossAttention (BiSSM-CA) module, which efficiently learns and retains the distinct styles of Chinese landscape paintings. Additionally, we introduce two latent space feature adjustment methods, Latent-AdaIN and Latent-WCT, to enhance style modulation during inference. Experiments demonstrate that CLPFusion produces more realistic and artistic Chinese landscape paintings than existing approaches, showcasing its effectiveness and uniqueness in the field.