Browse Articles

Gray Wolf Diet Composition in California’s Human-Dominated Landscape

Gray wolves have successfully recolonized much of their historical range on multiple continents, now occupying landscapes substantially altered during their absence. Wolf ecological function in human-modified landscapes remains poorly understood, fueling conflict concerns among rural stakeholders and conservation advocates over predation on economically valuable livestock and wildlife. Dietary studies can facilitate prediction of ecological impacts and help avoid outcomes that undermine human-wolf coexistence and conservation objectives. This study addresses a critical gap in dietary research of a recolonizing population of wolves in California, USA, representing the first investigation employing genetic identification of species and individuals combined with DNA metabarcoding for diet determination in scat samples. Our findings indicate that cattle dominated wolf diet, occurring in 72% of samples during the study period, compared to mule deer, which occurred in 45% of samples and small mammals that appeared in 51% of samples. Dietary biomass estimates corroborated cattle as the principal dietary component accounting for an average of 55%, while mule deer and small mammals each contributed considerably smaller proportions, 12% and 15%, respectively. These results revealed that wolves in California’s human-modified landscape primarily satisfy their caloric requirements through livestock consumption, underscoring the considerable challenges posed by coexistence of wolves and humans within the state.

Systems-level analysis predicts no autotrophy-linked protein-RNA interactions in Clostridium autoethanogenum

Abstract Burning fossil fuels drives climate change with detrimental consequences on humankind while waste accumulation from increasing consumption threatens biosustainability globally. Gas fermentation enables to recycle carbon oxides (CO and CO 2 ) from industrial waste gases and gasified waste into value-added products using gas-fermenting microbes, namely acetogens. However, our limited understanding of gene function and metabolic regulation is hindering rational engineering of acetogen cell factories. In this work, we aimed to identify genome-wide protein-RNA interactions contributing to autotrophy in the model-acetogen Clostridium autoethanogenum by combining steady-state chemostat cultivation, functional genomics, and computational methods. We first detected limited and uncoupled transcriptional and translational regulation between autotrophy and heterotrophy. Rigorous mapping of genome-wide transcriptional architecture revealed both differential usage and signal strength of transcriptional start and termination sites between genes and growth substrates. We then used computational tools to reconstruct protein-RNA interactions for differentially regulated genes, predicting 14 trans-acting regulatory RNA-binding proteins (RBPs) involved in post-transcriptional regulation but not linked to genes key for autotrophy. Most RBPs, one of which is translationally regulated, perform RNA modifications and regulate mRNA stability while others target translation-related genes. Our work provides valuable knowledge for metabolic engineering of acetogens and potentially contributes towards understanding primordial life on Earth.

Effects of node number of vine cuttings and planting methods on endogenous hormone contents at nodes and storage root formation in sweet potato

As a globally significant dual-purpose crop, sweet potato yield is regulated by the node number of vine cuttings and the planting method. Elucidating the effects of stem segment number and soil-entry method on endogenous hormones within sweet potato nodes and storage root formation holds significant scientific value, providing theoretical basis and technical support for optimizing planting techniques and enhancing yields. A two-factor split-plot design was implemented from 2022 to 2023 using the Yuhongxinshu 98 cultivar. Main plots tested vine cuttings with 6 (A1), 8 (A2), or 10 nodes (A3). Subplots compared slanted planting with exposed tip (B1), flat planting with exposed tip (B2), and flat planting without exposed tip (B3). The study investigated the effects of these treatments on growth hormone content at nodes, root morphology, and yield. Results indicated that increasing vine cutting node number to 8 and 10 significantly enhanced early adventitious root development, including root length, root diameter, and root tip number, and increased aboveground branching. Compared with 6-node cuttings, 8- and 10-node cuttings increased storage root yield by 23.43% and 30.03%, respectively. Compared with slanting planting with exposed tip, flat planting with exposed tip and flat planting without exposed tip increased storage root yield by 20.11% and 18.10%, respectively. At 70 DAT, corresponding to the early storage root enlargement stage, ABA and IAA contents in buried nodes increased with vine cutting node number. IAA content showed a node-dependent pattern and peaked at the fourth buried node, whereas ABA content was highest at the second buried node. Correlation analysis showed that ABA and IAA contents were associated with storage root yield, storage root number, and root morphological traits. Nodal analysis indicated that storage root formation was mainly concentrated in the upper buried nodes, especially the second buried node, while flat planting increased the contribution of multiple buried nodes to storage root formation. Overall, 8-node vine cuttings combined with flat planting, especially A2B2 and A2B3, showed better performance in coordinating early root establishment, nodal storage root formation, and yield improvement. These findings provide a practical basis for optimizing sweet potato planting techniques through improved vine cutting node number and planting method.

Non-equilibrium correlated electron dynamics in triangular molecular assemblies

Abstract Understanding and controlling charge states at the level of individual electrons in molecular assemblies is often hindered by system complexity and the lack of reliable theoretical framework to accurately model the underlying many-body non-equilibrium electron dynamics. To address this challenge, here we construct well-defined molecular trimers and hexamers of tetrabromo-tetraazapyrene molecules by scanning probe manipulation and show that the electron dynamics can be completely rationalized using the Anderson impurity model and master equation approach. Our analysis demonstrates that such a treatment, which goes beyond the standard single-particle picture, is essential for understanding this class of molecular systems. The model not only quantitatively reproduces all features visible in measured differential conductance maps, but also uncovers the mechanism for negative differential conductance, which arises from the non-equilibrium occupancy of collective charge states. The analysis also reveals that the charging rings are not necessarily associated with changes in the total charge of the cluster, but often originate from internal charge rearrangements between the sites. These findings provide insight into fundamental quantum many-body effects in strongly interacting molecular systems and open pathways for engineering electronic phases by controlling cluster topology and the electrostatic environment.

Beyond traditional outbreak investigation: Using genomic data for enhanced detection of COVID-19 disease clusters in Utah

Background Traditional disease surveillance, such as manual case investigation, was the primary method for identifying disease clusters during the COVID-19 pandemic. However, the pandemic also provides an opportunity to explore how genomic data can be used to improve cluster detection and response. While genomic data can complement traditional methods, guidelines are needed to integrate genomic data into real-time outbreak response. Methods Using binomial and multinomial logistic regression, we compared two methods of disease surveillance in Utah: genomic sequencing of COVID-19 cases and manual case investigation. We evaluated whether these two methods reached the same populations geographically and demographically. Next, we performed genomic clustering using SNP distance thresholds and a logit regression model to identify potential transmission clusters. We compared genomic clusters with epi-identified clusters, defined by manual case investigation, using cluster validation metrics (Adjusted Rand Index, VI), and by assessing biological plausibility (monophyly). Results The odds of a case being sequenced varied significantly by jurisdiction and race/ethnicity, with patients in several non-White groups being less likely to undergo sequencing. The genomic clustering methods produced clusters that were notably different from epi-identified clusters. Genomic methods, particularly the logit model, resulted in strong clusters based on metrics of cluster validation and biological plausibility. Analysis of specific epi-defined clusters revealed significant discordance with genomic data. Many large clusters were likely composed of multiple distinct genomic introductions, or contained cases that were not genomically linked. Conclusions Genomic data provides an advanced level of resolution for defining disease clusters compared to traditional epidemiological data. The disparities in sequencing coverage necessitate demographically and geographically diverse sampling strategies. Furthermore, it is essential to prioritize sequencing cases in a suspected cluster to maximize the impact of genomic surveillance. Integrating genomic data into epidemiologic investigation enables more precise cluster definitions, strengthening outbreak investigation and public health mitigation.

Dynamic control of laser driven electron acceleration in a photonic structure using programmable optical pulses

Abstract Progress in optical techniques has made precision control of the phase profile in optical pulses common and accessible in scientific laboratories. Carefully shaping the field profile of a laser pulse can be used to master the dynamics of electrons traveling in photonic accelerating structures, such as the ones obtained by precisely aligning two dielectric gratings. Here, we show that by applying a liquid-crystal mask to program the phase and amplitude of an infrared laser pulse in combination with a pulse front tilt scheme, it is possible to implement dynamic control on a laser accelerator. This results in a nearly limitless live tuning capability of the accelerator beam dynamics, allowing the demonstration of software-based correction of structure and optical front imperfections, implementation of transverse focusing schemes, control of the energy and charge of the output beam, and ultimately optimization of the interaction length, leading to measured energy gains of up to 0.55 MeV.

Explainable machine learning for the early differentiation of pediatric bronchopneumonia using routine laboratory parameters

Objective Early differentiation of pediatric acute respiratory infections in outpatient and emergency settings is often hindered by nonspecific clinical signs, leading to unnecessary empirical antibiotic use and avoidable radiation exposure. Therefore, this study aimed to develop and internally evaluate an early triage model to distinguish pediatric bronchopneumonia (BP) from uncomplicated upper respiratory tract infection using routine, cost-effective laboratory parameters analyzed with machine-learning algorithms. Methods This retrospective study consecutively enrolled 532 pediatric patients who presented with mild respiratory symptoms at their initial visit, comprising 218 in the BP group and 314 in the Upper Respiratory Tract Infection (URTI) group. Core laboratory indicators were selected using a dual-dimensionality reduction strategy that integrated the Least Absolute Shrinkage and Selection Operator (LASSO) regression with the Boruta algorithm. Seven machine learning classifiers were then constructed and compared based on the resulting feature matrix. The Shapley Additive Explanations (SHAP) framework was subsequently applied to interpret the nonlinear predictive mechanisms of the optimal model. Results This study constructed seven machine learning models: Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Lightweight Gradient Boosting Machine (LightGBM), Support Vector Machine (SVM), and Artificial Neural Network (ANN). In the internal validation set, the SVM model demonstrated favorable predictive performance, with an area under the curve (AUC) of 0.921 (95% confidence interval: 0.874–0.959). Based on Platt-scaled probability estimates, the SVM model showed the lowest Brier score among the evaluated models, with a Brier score of 0.112. Decision curve analysis confirmed this model’s positive net clinical benefit across a broad range of threshold probabilities. SHAP analysis further elucidated the nonlinear contribution weights of multiple conventional parameters at specific physiological thresholds. Conclusions The multidimensional SVM risk quantification model, based on nine routine laboratory parameters, provides an accurate and objective assessment of pediatric BP risk. This model holds significant potential for clinical translation as a noninvasive, cost-effective triage tool in emergency departments. Its application could effectively reduce unnecessary radiographic screening and excessive antibiotic use.

A single-cell atlas of porcine hematopoietic development

Association between depressive symptoms and balance disorders in older adults living in 12 high Andean communities

Background The prevalence of depressive symptoms and balance disorders may be higher in high-altitude Andean regions due to chronic hypoxia, neurochemical alterations, and limited access to health services. Since both conditions can coexist and contribute to functional decline and falls, their association is relevant. We aimed to estimate the association between depressive symptoms and balance disorders in older adults living in 12 high Andean communities. Methods We carried out a secondary analysis of data from a cross-sectional analytical study in older adults residing in 12 Peruvian high Andean communities during the period 2013–2020. The exposure variable was depressive symptoms (defined as a score greater than or equal to two on the five-item geriatric depression scale), while the outcome variable was balance disorders (defined by a functional reach test less than or equal to 20.32 cm). We constructed generalized linear models from Poisson family with link log and robust variances. We estimated crude (cPR) and adjusted (aPR) prevalence ratios with their respective 95% confidence intervals (95%CI). Results We analyzed 417 older adults; 61.1% (n = 255) were women, with a mean age of 73.2 ± 6.9 years. Additionally, 52.8% (n = 220) presented depressive symptoms, while 48.9% (n = 204) presented balance disorders. In the adjusted regression model, depressive symptoms were associated with a higher prevalence of balance disorders in older adults (aPR = 1.66; 95%CI: 1.28–2.15; p < 0.001). Conclusions Depressive symptoms were associated with a higher prevalence of balance disorders in older adults residing in the 12 high Andean communities. Future epidemiological studies with a larger sample size are needed to evaluate depressive symptoms and balance disorders to develop early screening programs in older adults to improve their quality of life and access to primary health care.

A NIR fluorotag reporter CETIF6a enables bright pan-tumor labeling and functional proteomic profiling

Abstract Tumor-seeking fluorescent dyes enable precise lesion localization by recognizing overexpressed receptors, providing a critical adjunctive technology for cancer histopathology. However, tumor heterogeneity and the poor understanding of targeting mechanisms limit their efficacy. Here we engineer CETIF6a, a click chemistry-compatible heptamethine cyanine dye, for multi-cancer targeting, intraoperative histopathology, and proteome-wide target identification. CETIF6a demonstrates margin delineation across multiple cancer types ( > 90% concordance with H&E staining). Quantitative proteomics reveals that the dye targets 5-15 times more tumor-specific proteins than in paracancerous tissues. Synergistic pan-cancer targeting is achieved through 491 conserved tumor-enriched proteins involved in ribosomal, proteasomal, and metabolic pathways, effectively overcoming heterogeneity. Mechanistic studies confirm that CETIF6a emits bright fluorescence upon covalent binding to targets via nucleophilic substitution at cysteine thiol residues within hydrophobic cavities. The modifiable scaffold of CETIF6a supports both intraoperative tumor diagnosis and functional targets profiling, providing a foundation for systematic probe optimization.

Serum albumin and blood urea as independent predictors of in-hospital mortality in hospitalized COVID-19 patients: A retrospective cohort study

Background While systemic inflammation is a hallmark of severe COVID-19, the prognostic value of metabolic and organ-functional markers remains under-explored. This study examined the prognostic value of serum albumin and blood urea at hospital admission in predicting in-hospital mortality among patients with COVID-19. Methods We performed a retrospective cohort study of 1,074 adult patients with laboratory-confirmed COVID-19 who were hospitalized in a tertiary care center. Patients’ demographic, clinical and laboratory data on admission were collected from electronic medical records. We conducted multivariable logistic regression analyses to determine independent predictors of in-hospital death after adjusting for possible confounders including age, sex, comorbidities and known inflammatory/coagulation markers. Missing laboratory variables were imputed by multiple imputation by chained equations. Area under the receiver Operating characteristics curve (AUC) was evaluated for model discrimination. Results In-hospital mortality occurred in (24.5%). Non-survivors had significantly lower serum albumin and higher blood urea levels at admission. After full adjustment, hypoalbuminemia remained independently associated with increased mortality risk. Each 10 mg/dL increase in urea was associated with a 14% increase in the odds of death (AOR = 1.14, 95% CI 1.08–1.20), whereas albumin demonstrated an independent protective association with mortality (AOR = 0.52, 95% CI 0.37–0.74). The final multivariable model demonstrated good discrimination (AUC = 0.86), indicating strong predictive ability. Conclusions Serum albumin and blood urea at hospital admission independently predict mortality in patients with COVID-19, even after accounting for inflammatory and coagulation markers. These findings suggest that markers of organ-functional reserve may provide prognostic information beyond traditional inflammatory markers in hospitalized patients with COVID-19.

Dynamic interactions between epithelial skin cells and a sensory cavity sculpt the growing olfactory orifice

Abstract During morphogenesis and in pathological conditions, gaps can form in the plane of epithelial barriers upon cellular forces that disrupt intercellular junctions. How the size of these epithelial holes further increases over time and what sets their shape remain poorly understood. Here we analyze the formation of the olfactory orifice (the nostril) in zebrafish, which opens and grows in the skin epithelium above a rosette of olfactory placode cells, allowing the sensory neurons to directly access odor cues. Using quantitative imaging and tissue-specific perturbations, we analyze the dynamic remodeling of skin cells allowing the expansion of the orifice edge. We identify the sensory cavity located in the center of the placodal rosette as a crucial player that sets the size of the growing epithelial hole in the skin. We further show that fine-tuning of actomyosin contractility within each tissue (skin and sensory cavity) exerts non-autonomous effects on the neighboring tissue, thereby shaping the nostril structure. This study uncovers dynamic cell behaviors and reciprocal tissue-tissue interplay that control the growth and shape of an epithelial hole in vivo.

Complex intervention programme to improve patient safety and facilitate deprescribing in frail older patients living at home (COFRAIL): A process evaluation of a cluster randomised controlled trial

Introduction Frailty is associated with negative health outcomes in geriatric patients. A large proportion of frail patients is affected by polypharmacy, which in turn may be a possible cause of frailty. The cluster randomised controlled COFRAIL trial investigated the effects of family conferences to improve the care of frail patients. Deprescribing and communication about prioritising health goals between patients, relatives and general practitioners were key components of the intervention. An accompanying process evaluation was conducted to investigate whether the study intervention was implemented as intended and to describe the experiences of the target groups. Methods The process evaluation took place between February 2019 and October 2021 and followed international guidelines. Process parameters were collected using study documentation, standardised questionnaires and guided telephone interviews with a convenience sample of patients, relatives and general practitioners. Quantitative data were analysed descriptively, qualitative data through content analysis. Results Almost all general practitioners in the intervention group completed both mandatory trainings. Overall, 68% of the patients took part in all three planned family conferences, 85% in at least two. Patients, relatives and general practitioners reported positive experiences with the family conferences. Patients and relatives felt involved in the decision-making process and had predominantly no concerns when medication was discontinued. Only a few general practitioners considered the implementation of family conferences in regular care to be impractical. A supportive pharmacological hotline was not utilised frequently by the general practitioners. Due to the SARS-CoV-2 pandemic, some study procedures were adapted but did not have any negative impact. Conclusions While the COFRAIL study did not achieve its primary goals, we identified no major barriers in the implementation of the intervention programme. Overall, the target groups experienced the approach of family conferences positively. However, the process evaluation provides valuable lessons for designing and implementing similar interventions in the future.

A portable Cas6f-based system for multiplex translational repression in bacteria

Development and psychometric evaluation of the Spanish Cooking Self-Efficacy Questionnaire (SCSEQ) for Spanish University Students

The transition to university is a critical life stage characterized by increased autonomy, identity exploration, and new social and environmental influences. During this period, university students often exhibit low adherence to dietary guidelines. Among the determinants influencing healthy eating, cooking self-efficacy, the central construct of Social Cognitive Theory (SCT), is consistently associated with improved diet quality and is a frequent target of health interventions. However, no validated instrument exists to assess this construct among university students in Spain. Therefore, the goal was to develop and provide preliminary evidence of the Spanish Cooking Self-Efficacy Questionnaire (SCSEQ), a concise SCT-based instrument tailored to Mediterranean university settings. A 32-item questionnaire was developed through a review of existing instruments assessing cooking self-efficacy. Face validity was evaluated with Spanish food and nutrition experts (n = 12) to assess the clarity and pertinence of the initial items. The revised Spanish Cooking Self-Efficacy Questionnaire (SCSEQ) was then pilot-tested with Spanish university students (n = 73) from four Catalan universities. Exploratory factor analysis (EFA) was conducted to identify the underlying factor structure and detect problematic items. Internal consistency reliability was assessed using McDonald’s ω, and test–retest reliability over a two-week interval was evaluated using Pearson correlations. Face validity indicated overall clarity and adequacy. Four items were excluded and recombined, two items were added, and nine items were rewritten based on experts’ feedback. After pilot testing, the questionnaire overall demonstrated high internal consistency (ω = 0.9). Items were reviewed based on factor loadings, item redundancy, theoretical relevance, and their contribution to scale-level internal consistency. EFA suggested a two-factor structure with good internal consistency (ω = 0.88 and ω = 0.82) and test–retest reliability (ICC = 0.91, 95% CI [0.80, 0.96]). Three items with weak loadings were excluded. The final version consisted of 25 items and 2 subscales. The SCSEQ showed favorable preliminary psychometric properties.

Sample-efficient fine-tuning with textual prompts for time series forecasting

Time series forecasting models often face challenges in cross-domain fine-tuning, such as high training costs and limited adaptability. To address these issues, we propose a Cue-driven Feature Fusion Network (CFF-Net), which combines semantic cues from textual prompts with numerical time series features for parameter-efficient adaptation. The main idea is to use the semantic representation ability of large language models to provide auxiliary guidance, while dynamically modulating numerical predictions through scale-and-shift operations. Specifically, CFF-Net includes three main components. First, the Semantic Prompt Encoding Module (SPM) transforms numerical sequences into temporally relevant natural language descriptions, which are processed by GPT-2 to extract semantic representations. Second, the Dynamic Semantic Modulation Module (DSM) maps these semantic representations into learnable scaling ( γ ) and shifting ( β ) factors through a multi-layer perceptron, enabling modulation of PatchTST predictions within the Scale-and-Shift Feature (SSF) mechanism. Finally, a warm-start strategy is used to stabilize semantic integration during training. Experimental results on three public datasets and the TCTS dataset show that CFF-Net achieves lower errors than PatchTST in many settings, although the improvements are not uniform across all datasets and metrics. For example, on the Weather dataset, CFF-Net reduces MSE by 12.50% and 11.88% under the 30% and 20% training-sample settings, respectively. On the TCTS dataset, the corresponding MSE reductions are 5.83% and 6.60%. These results suggest that semantic prompt guidance can improve forecasting performance in several limited-data scenarios while keeping most backbone parameters fixed.

Effects of season and antler growth on carcass traits and meat quality in wild male red deer harvested by selective stalking

Seasonal variation in temperate ecosystems influences food availability and the energetic demands associated with thermoregulation and reproduction, shaping physiological traits such as meat composition in wild animals. However, the specific effects of seasonality on meat quality in wild red deer remain insufficiently understood. This study evaluated how seasonality —represented by hunting month (September, January, April, and June)— affects the quality of the Longissimus thoracis et lumborum muscle in 32 adult wild male red deer harvested under minimal stress conditions. We hypothesized that seasonal fluctuations in diet and physiological requirements, particularly those linked to antler growth, would generate measurable differences in carcass traits and meat composition. Deer hunted in September (late summer) showed the highest carcass weight and yield, whereas the lowest values occurred in January (mid-winter) and June (early summer; p < 0.001). Intramuscular fat content was also greatest in September, with significant differences compared to April (p < 0.05). Additional seasonal effects were detected for pH at 72 h post-mortem (pH₇₂), shear force, and fatty acid profile (p < 0.05), indicating that muscle metabolism and textural properties vary across the year. Mineral composition exhibited a distinct pattern, suggesting a stronger association with skeletal mobilization during antler growth than with dietary changes. Mineral composition patterns were consistent with seasonal skeletal mobilization during antler growth: Ca and Mg were higher in April; Fe and Zn were lower in April and June. Overall, these findings demonstrate that both season and antler-growth effort affected meat quality and composition in wild male red deer, providing new insights into nutritional ecology and game meat quality.

Beaver in tidal habitat: Examples from the Pacific Northwest

Beaver are typically considered fluvial or lacustrine animals, often converting lotic habitats into lentic ones with their dams. This results in extensive changes in ecosystem structure and processes so that beaver are considered the quintessential ecosystem engineer. Here I broaden our appreciation of the adaptability of beaver by describing their widespread presence in tidal river deltas and estuaries of the Pacific Northwest (coastal British Columbia, Washington, and Oregon), where tides can range between 1.5 and 5.0 m. These observations expand the known habitat distribution of beaver and invite investigation of the ecosystem consequences of beaver in tidal wetlands. In these oligohaline to fresh tidal systems, channel profile surveys with real-time kinematic (RTK)-GPS show that beaver dams are typically flooded on higher high tides, only impounding water at low tide to allow beaver movement during this time. Tidal beaver dam density per km was more than twice that reported in the fluvial literature, while mean dam head was about 80% and mean dam height about 60% of fluvial dams. Low-tide beaver pool depths were 75% of fluvial beaver ponds, while pool areas were 67% of their fluvial counterparts. The comparable beaver dam metrics between fluvial and tidal systems suggests their role in tidal ecosystems may be comparably significant. Inspection of Google Earth aerial photographs back to 1990 indicated that tidal beaver dams can persist for at least 35 years, spanning several generations of beaver. Given the ecosystem importance of beaver in fluvial and lacustrine habitats, better understanding of the distribution and ecosystem role of beaver in tidal wetland habitat, and the geometry of their dams and low-tide pools, would likely allow more effective restoration of estuarine habitat that is critical to a variety of fish and wildlife, including threatened species such as Chinook and coho salmon.

Dynamic changes in In vitro rumen fermentation, nutrient degradation, and microbial communities of fermentation inoculant-treated licorice stem and leaf silage under different dry matter contents

This study aimed to explore the effects of semi-dry licorice ( Glycyrrhiza uralensis ) stem and leaf silage on dynamic changes in dairy cow in vitro rumen fermentation parameters, nutrient degradation rates, and microbial community composition. Silage substrates with four dry matter (DM) contents (45%, 50%, 55%, 60%; L45, L50, L55, L60) were prepared by adjusting moisture with additives, with 5 replicates per group per time point. In vitro fermentation lasted 72 h, with cumulative gas production (GP), fermentation parameters, microbial composition, and nutrient degradation rates measured at 0, 3, 12, 24, 72 h. Principal component analysis (PCA) and gray relational analysis (GRA) were used for comprehensive evaluation. Results showed that L55 had significantly higher 72-h GP, microbial crude protein (MCP), and nutrient degradation rates ( P  < 0.05), and lower pH at 3 and 72 h ( P  < 0.05). It also had higher total volatile fatty acid (TVFA) and acetic acid (AA) at 12, 24 h ( P  < 0.05), the highest proportion of unique OTUs at 3, 72 h, and superior α-diversity indices (ACE, Chao1) compared to L45 (P < 0.05). L55 showed dominant early abundances of functional genera (e.g., Prevotella ) and higher metabolic pathway proportions ( P  < 0.05 vs. L60), ranking first in PCA and GRA. Conclusion: 55% DM in licorice semi-dry silage optimizes in vitro rumen fermentation and microbial composition, presenting the best feeding value.

Evidence-based sustainable business model design for agrifood side-stream biostimulants

Circular bioeconomy initiatives often valorise food-processing residues into bio-based inputs, yet environmental gains depend on how technologies are implemented and governed across multiple actors. This study evaluates a pathway that converts leek processing residues into a bioactive extract used as a plant biostimulant in greenhouse lettuce production and develops an evidence-based procedure to design a sustainable business model around it. We conducted a cradle-to-gate Life-Cycle Assessment (LCA) of leek extract production and a comparative LCA of lettuce cultivation for a baseline system and an alternative system applying the extract, using the Environmental Footprint (EF) 3.1 midpoint indicators. The alternative scenario improved most indicators, with consistent reductions in acidification, eutrophication and freshwater ecotoxicity, while selected trade-offs were associated with the upstream extract supply chain, particularly in resource-related and process-sensitive impact categories. We then embedded these multi-impact hotspot signals into an ecosystem-oriented Sustainable Business Model Canvas (SBMC) and an experimentation logic, specifying who must act on each hotspot (processors, extract producers, labs, farmers) and how mitigation and verification can be financed through value capture (product-service bundles, monitoring, and conditional sustainability claims). The approach links LCA evidence to actionable business model choices, supporting credible scaling of residue-based agrifood innovations.