Journals
2026 EN
Kizmaz Kerem · Emire Zuhal · Ugras Serpil
There are many enzymes used in various industries, and the application areas of enzymes are increasing day by day. Among these enzymes, cellulase enzymes, which are of industrial importance, are used in many industrial sectors such as paper pulp, textiles, and food. In this study, bacteria isolated from bovine feces that produce cellulase enzyme were screened. Then, a bacterium identified as a producer, coded as UR4. The isolated bacterium was identified as Cellvibrio polysaccharolyticus ( Cp -UR4). Thus, for the first time in the literature, the Cp -UR4 was used for cellulase production in this study. The optimum production conditions of the enzyme were determined to be 37 °C, pH 7.0, and 72 h of incubation. The enzyme working conditions were determined to be at 45 °C and pH 7.0. The Vmax and Km values for enzyme kinetics were found to be 6.41 U/mL and 12.53 mM, respectively. Ammonium sulfate method was used for partial purification of the enzyme in the final stage. Native-PAGE and zymogram analysis were performed to determine the molecular weight of the partially purified enzyme. It was determined that the cellulase formed aggregates larger than 130 kDa. The enzyme was applied to denim fabric, and the results showed that it performed better in de-fuzzing, bio-polishing, and color removal compared to the reference enzyme. Thus, this study suggests that cellulase obtained for the first time from Cp-UR4, isolated from bovine feces, can be used for bio-polishing, color removal, and de-fuzzing processes in the textile industry.
Journals
2026 EN
Liu Songlin · Yüce Erkan · Solhi Mehdi
+2 more
Although enthusiasm is an important characteristic of teachers’ psychological growth, scant research attention has been given to English as a foreign language (EFL) teachers’ enthusiasm and its outcomes in the field of foreign language education. In addition, in spite of the criticality of foreign language teaching enjoyment (FLTE), little is known about its role in mediating the effect of enthusiasm on job satisfaction. Therefore, the current study aimed to investigate whether Turkish EFL teachers’ FLTE could mediate the relationship between their enthusiasm and job satisfaction. The participants ( N = 314) completed three scales and the data were analysed through structural equation modeling (SEM). The results of the mediation analysis revealed that a. EFL teachers’ enthusiasm significantly predicted their job satisfaction and b. FLTE partially mediated the relationship between EFL teachers’ enthusiasm and job satisfaction. Therefore, enthusiasm would, directly and indirectly, through teaching enjoyment, result in teaching satisfaction. The findings of this study highlighted the broadening capacity of teacher enthusiasm in offering positive outcomes. Finally, implications and suggestions for future research were offered.
Journals
2026 EN
Neofytou Andreas · Liu Bin · Akartunalı Kerem
This study utilizes unsupervised machine learning to improve the performance of robust optimization through construction of compact uncertainty sets. Robust optimization aims to find solutions that are resilient to uncertainties and variations in input parameters, allowing decision-makers to make informed decisions in the face of uncertain conditions. However, traditional optimization approaches often assume known and fixed datasets, which fails to reflect the inherent uncertainties present in real-world problems. Accurate construction of compact and reliable data-driven uncertainty sets is a critical challenge that directly impacts the effectiveness of robust optimization. To address this challenge, we propose a Dirichlet process mixture model for clustering to construct a data-driven uncertainty set suitable for robust optimization problems, allowing for more accurate uncertainty modelling. This data-driven uncertainty set is constructed by intersecting the$ l_{1} $l 1and$ l_{ } $l ∞norms for each predicted cluster and then merging these multiple basic convex uncertainty sets to create a comprehensive representation. This approach results in uncertainty sets based on clustered data that flexibly capture a compact region of uncertainty in a nonparametric manner. An innovative aspect is the introduction of statistical bootstrap to obtain a more robust clustering solution and outcome. The proposed method is applied to production planning problems and a comparative analysis with existing approaches highlights its advantages. Our method demonstrates effectiveness in improving the accuracy and robustness of solutions in robust optimization by constructing more compact uncertainty sets.
Journals
2026 EN
Istif Inci Elçin · Altıntop Apak Kerem · Özgür Baklacıoğlu Nurcan
+1 more
This research elaborates the nexus of disability-migration-education through a multidisciplinary investigation of intersectional barriers faced by Syrian youth with disabilities during their access to, and study at higher education (HE) institutions in Turkey. HE inclusivity for Syrian students with disabilities (SSD) is analysed in terms of the main administrative, societal, and structural barriers. The methodological design of this study applies a human rights-based approach to education, migration and disability in company with participatory action research methodology. Findings are based on data collected through semi-structured interviews, participant observations, institutional visits, and multi-stakeholder workshops addressing disability and migration-based experiences of SSD in 2018 and 2019 in Istanbul, Şanlıurfa and Gaziantep. Information and experiences are collected from SSD, NGOs in the field of migration and disability, governmental bodies, and HE institutions. Our findings showed that various legal, administrative and cultural barriers, and structural inequalities make access to HE more difficult for refugees with disabilities. Despite these, SSD have success stories in accessing HE as a result of their hard work and courage. This study provides first-hand bottom-up intersectional research about the situation of SSD in Turkey, host to the largest refugee population in the world.
Journals
2026 EN
Özoflu Melek Aylin · Yiğit Uyar Merve Hazer · Altıntop Apak Kerem
+1 more
Although Türkiye is the largest refugee-hosting country in the world, the Zafer Party’s (ZP) (English: Victory Party) discourse on migration as the pioneer representative of the European-style anti-immigrant right party example has been under-studied. To address this gap, this research examines the discursive practices of the ZP through the lens of populist securitization. It focuses on revealing how the ZP employs a people-centred appeal to securitization, explaining how the people are located within its populist discourse as a threatened object of reference and actor. In this way, it empirically contributes to the burgeoning literature bridging between populism and securitization theory. To this end, it conducts an extensive qualitative frame analysis of the party’s manifesto, programme, press releases, public speeches uttered by party leader Ümit Özdağ. The time frame of the research is designated as starting from the establishment of the ZP, i.e. 26 August 2021, to the Presidential and Parliamentary Elections in Türkiye, i.e. 14 May 2023. The paper finds out that ZP combines populist and nationalist appeals by creating vertical (down/up) and horizontal (in/out) antagonisms, exhibiting parallels with similar tendencies in its European counterparts.
Journals
2026 EN
Turan Ferruh · Zeren Ertugrul · Basoglu Muhammed Fatih
+2 more
Porous laminated composite shells are increasingly critical for lightweight structural applications where achieving high stiffness-to-weight ratios is mandatory. This study comprehensively examines the coupled effects of porosity distribution, foundation orthotropy, shell curvature, lamination sequence, and material orthotropy on the fundamental frequency characteristics of orthotropic laminated doubly curved shallow shells. The governing motion equations are formulated using a higher-order shear deformation theory (HSDT) and solved via the Galerkin method, incorporating the interaction with an orthotropic Pasternak foundation. Numerical results reveal that the orthotropic foundation markedly enhances the fundamental frequency by up to 33% through directional shear coupling and lateral restraint. While porosity generally reduces frequencies, surface-stiff (NUDP3) and uniform (UDP) patterns interact most effectively with the foundation to mitigate stiffness loss. Notably, the orthotropic support reduces the frequency discrepancy between spherical and hyperbolic shells by nearly 25%. However, increasing the in-plane orthotropy ratio diminishes the influence of the Pasternak layer by 25–33%. These findings highlight that the integrated optimization of porosity distribution, lamination configuration, and foundation anisotropy is essential for maximizing the dynamic performance of next-generation composite shell structures.
Journals
2026 EN
Tahan Kerem · Cayrier Alexia · Baratgin Jean
+1 more
The objective of the present study was to evaluate the interest of an assistance robot to help caregivers manage the activities of daily living of institutionalized elderly people with Alzheimer’s disease. Twenty-three institutionalized persons (60% women; average age 89; average MMSE score of 20.8) with Alzheimer Disease (AD) were recruited and invited to participate in prospective memory exercise sessions, conducted either by a caregiver or by a robot (assisted by a caregiver). They were divided into two groups equivalent in age, level of education and MMSE score. In addition, the sessions were recorded in order to compare the interaction behaviors of the 2 groups, using a validated observation grid. The results showed that: 1) prospective memory tasks are better performed when offered by the caregiver; 2) when strong help linked to the recovery index is provided to perform the tasks, the robot or caregiver no longer show significant differences; 3) participants interact more with the caregiver than with the robot. Our results confirm that the use of companion robots is a promising way to help caregivers manage the daily activities of people with Alzheimer’s. However, to optimize this assistance, further investigations should be conducted to improve the fluidity of interactions between the patient and the robot.
Resource
2026 EN
Kerem Gencer
Acute kidney injury (AKI) affects up to 50% of critically ill patients and is independently associated with increased in-hospital mortality. Early, accurate identification of at-risk patients is critical yet remains challenging due to the heterogeneous presentation and multifactorial etiology of AKI in the intensive care unit (ICU). This paper presents AKI-PREDICT, a retrieval-augmented generation (RAG)-augmented clinical decision support framework that integrates gradient-boosted machine learning with SHAP-based explainability and biomedical evidence retrieval for early AKI prediction. The system is trained and evaluated on MIMIC-IV v3.1, a large real-world ICU database comprising 29,454 adult patients after KDIGO-aligned cohort construction. A strict temporal design confines all predictive features to the first 24 hours of ICU admission, with the AKI outcome window defined as hours 24–72, ensuring complete elimination of data leakage. The primary XGBoost model achieves an AUROC of 0.806 and AUPRC of 0.781 on the held-out test set, outperforming LightGBM (AUROC 0.804) and Random Forest (AUROC 0.786). A secondary multi-class XGBoost model performs KDIGO staging (Stage 0–3), achieving 68.4% overall accuracy. SHAP analysis identifies urine output rate (mL/kg/h, 0–24h), maximum serum creatinine, patient weight, and mechanical ventilation as the principal predictors of AKI onset. An ablation study demonstrates that urine output features contribute the largest single-group performance gain (ΔAUROC = 0.033), followed by laboratory values (ΔAUROC = 0.022). The RAG component retrieves semantically relevant excerpts from the KDIGO 2012 Clinical Practice Guideline and PubMed literature using SHAP-driven query construction and BioBERT-based embeddings indexed with FAISS, generating patient-specific, evidence-grounded clinical reports. Subgroup analysis confirms consistent discriminative performance across CKD, sepsis, vasopressor exposure, and age strata. AKI-PREDICT provides a fully reproducible, open-source pipeline linking prediction, explanation, and guideline-grounded evidence in a single workflow.
Journals
2026 EN
Wronski MarieLouis · Boutin Regine · Sailer Clara O.
+16 more
Abstract Background Exogenous oxytocin (OT) reduces reward‐driven food intake. Less is known about endogenous OT and eating behaviour. Preclinical studies suggest peripheral OT levels may reflect the opposite of central appetite‐related OT activity. In healthy females, circulating OT declines after eating. Whether this pattern exists in obesity and how endogenous OT relates to hedonic eating remains unclear. We hypothesised that OT would decrease postprandially in adults with obesity and that higher OT exposure, reflecting lower central OT signalling, would correlate with greater reward‐driven caloric intake. Methods Sixty‐one adults with obesity (56% female; age [mean ± SE] 33.55 ± 0.81 years; BMI 36.77 ± 0.62 kg/m 2 ) consumed a standardised meal following an overnight fast. OT was measured in peripheral blood pre‐meal and 30, 60, and 120 min post‐meal. Area under the curve (AUC) was calculated to capture OT exposure. Postprandial hedonic eating drive was assessed via visual analogue scales (VAS) and Cookie Taste Test (CTT). Meal‐related OT dynamics were analysed using linear mixed effects models, relationships between OT exposure and hedonic eating drive were examined with linear regression and mediation analyses. Results OT levels did not change in response to the meal. Greater OT AUC was associated with reduced postprandial satisfaction ( p = 0.008, d = 1.00) and higher CTT caloric intake ( p = 0.036, d = 0.62). The relationship between OT AUC and CTT caloric intake was mediated by OT's effect on postprandial satisfaction ( p = 0.014, proportion mediated = 53.28%). Conclusions In obesity, OT levels did not change postprandially. Greater OT exposure was linked to lower satisfaction and increased hedonic eating, suggesting dysregulated OT signalling in obesity potentially contributing to overeating.
Journals
2026 EN
Wilson Lauren · Zhao Zhenxiang · Divino Victoria
+4 more
Abstract Aims To assess the real‐world effectiveness of semaglutide versus tirzepatide in reducing major adverse cardiovascular events (MACE) among patients with overweight/obesity and established atherosclerotic cardiovascular disease (ASCVD) without diabetes in an insured US population. Materials and Methods This retrospective, observational cohort study used Komodo Research Data and included patients ≥45 years of age with overweight/obesity and ≥1 claim for myocardial infarction (MI), ischemic stroke, or peripheral artery disease first treated with semaglutide or tirzepatide between 13/5/2022–31/1/2025. Propensity score matching was used to balance key baseline characteristics between cohorts. Primary outcomes included revised 3‐point MACE (rMACE‐3: MI, stroke, all‐cause mortality) and revised 5‐point MACE (rMACE‐5: rMACE‐3, coronary revascularization, hospitalisation for heart failure). Cox proportional hazard models were used to compare time to first event for study outcomes. A secondary per‐protocol analysis was conducted where patients were censored at treatment discontinuation (gap in therapy >30 days). Results 10 625 patients were included in each matched cohort. Semaglutide was associated with statistically significant 29% (hazard ratio [HR] 0.71; p = 0.046) and 22% (HR 0.78; p = 0.040) reductions in the risk of rMACE‐3 and rMACE‐5, respectively, compared with tirzepatide. In the per‐protocol analysis, semaglutide continued to be associated with a significantly lower risk of rMACE‐3 (HR 0.43; p = 0.005) and rMACE‐5 (HR 0.57; p = 0.003) compared with tirzepatide. Conclusions This real‐world analysis of a large US claims database shows semaglutide was associated with early and significantly greater reductions in the risk of rMACE‐3 and rMACE‐5 versus tirzepatide among patients with overweight or obesity and ASCVD but without diabetes.