WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7150
Browse
7671 results
Search Results
Article XGBoost and LightGBM for Spectral Authentication of Pomegranate Juice Concentrate: A Comparison with Classical Chemometric Models(Springer, 2026) Yasa, Irem; Ilgen, Elif; Uyar, Seda; Ozen, Banu; Cavdaroglu, Cagri; Uzum, NazliEnsuring the authenticity of fruit juice products is a growing concern, particularly for high-value pomegranate juice concentrate (PJC), which is vulnerable to economically motivated adulteration. This study evaluated Fourier-transform infrared (FTIR), UV-visible (UV-Vis), and fluorescence spectroscopy to detect adulteration of PJC with sugar syrups (invert and liquid) and fruit juice concentrates (apple and grape). A total of 166 authentic and adulterated samples were analyzed by each technique, and the resulting spectra were modeled using chemometric (PLS-DA, OPLS-DA, PLS, OPLS) and gradient boosting algorithms (XGBoost, LightGBM). Spectral analyses showed clear differences between authentic and adulterated samples. A sequential multi-level classification strategy yielded high diagnostic accuracy, with FTIR-based models giving the best performance and validation correct classification rates above 97%. For adulterant level prediction, FTIR-based PLS and XGBoost models achieved R2 values above 0.97 and RPD values above 7, indicating strong quantitative capability. UV-Vis and fluorescence models provided moderate performance, with fluorescence models showing the greatest variability. Overall, the findings highlight the superior performance of FTIR spectroscopy, particularly when coupled with gradient boosting, and support its use with robust modeling tools as a rapid, reliable approach for routine juice authenticity assessment in industrial and regulatory settings for screening and confirmation of adulteration.Article Stress-Triggered Cell Elimination Stabilizes Resource-Limited Cell Collectives: An Agent-Based Model(Elsevier Sci Ltd, 2026) Korkmaz, BertanMitochondria, long regarded as the cell's powerhouses, also serve as intracellular quality-control modules that promote the elimination of damaged or proliferatively dysregulated cells. Alongside their energetic benefits, mitochondrial apoptosis regulation may have helped shape the earliest steps toward stable multicellular organization. An agent-based modeling framework is developed to isolate and quantify the evolutionary impact of a mitochondrial death-surveillance system under nutrient stress, independently of its energetic contribution. The model tracks two cell types-normal prokaryotic cells (NPCs) and dysfunctional prokaryotic cells (DPCs)-each characterized by three dynamic state variables: damage load, proliferation drive and energy deficit. Mitochondrial surveillance is formalized as a simplified rule that detects stress and eliminates dysfunctional cells when a critical dysregulation boundary is crossed. Across 1200 stochastic simulations spanning variable nutrient regimes and inoculum sizes, mitochondrial surveillance prolonged median colony-collapse time by approximately 17%-18% across both nutrient regimes. The results indicate that a mitochondria-linked, intrinsic apoptosis-like gate can stabilize simple microbial groups independently of bioenergetic benefit, and is consistent with the view that such quality-control mechanisms provided a selective advantage toward robust multicellular organization. The framework also provides a conceptual basis for viewing mitochondrial transplantation as a means to restore intracellular fate-decision control, rather than solely to augment energy supply, in settings such as aging and cancer.Article When Does Instruction Tuning Work in Biomedical NLP? A Structural, Task-Aware, and Safety-Critical Evaluation of Large Language Models(IEEE-Inst Electrical Electronics Engineers Inc, 2026) Nasution, Arbi Haza; Celikten, Tugba; Onan, AytugInstruction tuning has emerged as an effective mechanism for adapting large language models (LLMs) to biomedical natural language processing (BioNLP), enabling improved task generalization with reduced reliance on extensive domain-specific pretraining. However, empirical evidence shows that its effectiveness varies substantially across biomedical tasks, and the factors governing these variations remain insufficiently understood. In particular, prior work often treats instruction tuning as a uniform adaptation strategy, without systematically examining the roles of task structure, instruction design, and robustness. In this work, we present a comprehensive, structure-aware and task-aware evaluation of instruction tuning for biomedical NLP. We organize biomedical tasks according to output granularity and rigidity, and analyze how instruction structures interact with token-, sentence-, and document-level tasks across multiple model families. To move beyond performance-centric evaluation, we introduce diagnostic measures that capture instruction sensitivity, format violations, and safety-related failure modes. Our results reveal a consistent relationship between output rigidity and instruction tuning gain: tasks with flexible output spaces benefit more from instruction tuning than schema-rigid tasks. We further show that instruction sensitivity varies with task granularity and instruction structure, while robustness and safety considerations expose trade-offs, and that robustness and safety considerations expose trade-offs that are not reflected in standard accuracy metrics alone. Based on these findings, we derive practical design guidelines for instruction construction, model selection, and evaluation in biomedical settings. This study reframes instruction tuning as a structured design problem rather than a one-size-fits-all approach.Article Transfer of Development Rights for Agricultural Land Protection in Izmir’s Periphery: A Case Study in Torbalı †(MDPI, 2026) Akbudak, Hacer; Akpinar, FigenSince the 1950s, Türkiye has experienced rapid urbanization and urban expansion followed by continuous planning initiatives, yet these efforts have resulted in significant land degradation and unsustainable urban sprawl. As a remedy, legislators and administrations are increasingly turning to the use of transferable development rights (TDRs), which have arisen as an innovative land readjustment tool and have recently been incorporated into the spatial planning system. This paper examines the effectiveness of TDRs by analyzing the legislative framework and operational rationale of the Turkish model through a hypothetical scenario, while also considering the institutional restrictions that could limit its usefulness as a sustainable planning instrument. By contrasting the scenario model with the framework recently developed through legal reform, this study employs the success factors of TDRs from the literature to assess the effectiveness of the tool integrated into the spatial planning system. Since the new legislation that forms the basis of the TDR model was passed in late 2024, empirical data on completed transactions is currently unavailable, and hence, the analysis used a hypothetical what-if-case scenario model in the local context, the Muratbey-Torbalı district of Izmir. To clarify the rationale behind incorporating the TDR into the planning system, we will first examine its conceptual development within the national legislation, followed by a critical evaluation of the TDR model as established by the recent amendment. Secondly, the study will present a hypothetical TDR model that incorporates the essential components ex-ante and offer guidance for conducting a market-based evaluation of TDRs, considering factors influencing agricultural market values and related standards. The findings demonstrate that there is a lack of legal clarity that stresses the program's holistic design with sending and receiving locations or TDR-allocation rates. The implementation regulation is challenging to ascertain how transfers impact land use justice, social benefit, and the public interest.Article Performance Evaluations of Indoor PV Devices under Various Light Illuminations(Pergamon-Elsevier Science Ltd, 2026) Varlikli, Canan; Ilhan, Hatice; Bazkir, OzcanThe power conversion efficiency of photovoltaic (PV) technologies strongly depends on incoming light spectrum and intensity. This study comparatively evaluated organic photovoltaics (OPV), dye-sensitized solar cells (DSSC), perovskite solar cells (PSC), amorphous silicon (a-Si) and crystalline silicon (c-Si) solar cells under four representative light sources, i.e. AM 1.5 spectrum, CIE Illuminant A, white LED and fluorescent lighting (FL), over a wide range of indoor illumination levels. In addition to electrical characterization, the optical properties of the light sources were analysed and the device spectral response was assessed. The results indicate that PV behaviour is strongly governed by the spectral distribution of the light source and the intrinsic optoelectronic properties of each device. PSC exhibits the highest overall performance due to its broad visible absorption and efficient charge transport. OPV maintains nearly constant fill factor and delivers relatively uniform efficiencies under all four spectra, highlighting their robustness under low-light conditions. DSSC shows excellent spectral compatibility with FL, leading to stable operation at moderate and high illumination levels. The a-Si device demonstrates moderate yet stable performance across all spectra, indicating balanced spectral tolerance. In contrast, c-Si displays pronounced spectral sensitivity, performing well under AM 1.5 and CIE Illuminant A but showing significant efficiency losses under LED and FL due to spectral mismatch at low photon flux. This study provides a comprehensive comparison of PV technologies under diverse illumination conditions and highlights the importance of combined optical and electrical characterization for reliable indoor PV performance evaluation.Article Quantum Key Distribution Using hBN Single-Photon Emitters at a 40 MHz Clock Rate(IOP Publishing Ltd, 2026) Aglarci, Furkan; Ates, Serkan; Pousa, Roberto G.; Oi, Daniel K. L.; Tapsin, Omer S.; Gundogan, MustafaRoom-temperature (RT) solid-state quantum emitters are essential for building practical and scalable quantum communication systems, yet their application has been critically hindered by the slow operational speeds of corresponding modulation technologies. In this work, we overcome this key performance bottleneck. We demonstrate a quantum key distribution (QKD) system using a single defect in hexagonal boron nitride (hBN) with dynamic polarization encoding at a 40 MHz clock rate, an order of magnitude faster than previous demonstrations with similar sources. Implementing the B92 protocol, our system yields a secure key rate of 7 kbps in the finite-key regime with a quantum bit error rate of 6.49%, establishing a new performance benchmark for RT single-photon QKD. Furthermore, to chart a path beyond the limits of direct transmission, we present the first quantitative performance analysis of hBN spin-defects as quantum repeater nodes. Overall, our high-speed experimental demonstration, supported by a foundational analysis of the system architecture, suggests that hBN defects represent a promising and technically feasible platform for scalable, quantum communication.Article One-Pot Green Synthesis of Amino Acid-Capped Gold Nanoparticles for Selective Sensing of Cyanide and Heavy Metals(Amer Chemical Soc, 2026) Arslan-Yildiz, Ahu; Girgin, Beylem; Sozmen, Alper BaranIn this study, 20 amino acids were utilized both as reducing and capping agents in a one-pot green synthesis of gold nanoparticles (GNPs) to be used in sensor applications for water, environment, and food monitoring. Tyrosine, tryptophan, valine, serine, phenylalanine, arginine, glutamic acid, and cysteine proved to be more suitable under the tested conditions, compared to the other amino acids, considering their colloidal stability. Amino acid-capped GNPs (AAGNPs) were then characterized in terms of absorbance spectrum, size, zeta potential, polydispersity index, geometry, and atomic content. After characterization, synthesized AAGNPs were utilized in sensory applications for Cyanide (CN-) and heavy metal (Al3+, Cu2+, and Fe3+) detection. The synthesized AAGNPs exhibited promising CN- detection capability that is comparable to conventionally synthesized GNPs via the Turkevich method. Besides, ArgGNPs, GluGNPs, and CysGNPs exhibited distinct sensing behaviors, reflecting differences in surface chemistry and interaction mechanisms. Sensing platforms that utilized PheGNPs, TrpGNPs, and TyrGNPs showed detection limits in the range of 0.3-0.7 mu M. Amino acid capping of GNPs imparted differential recognition capability toward various heavy metal ions, where detection limits were calculated for various AAGNPs as 0.27 mM for Cu2+ with CysGNPs, 0.25 mM for Al3+ with GluGNPs, and 0.44 mM for Fe3+ with SerGNPs. This phenomenon shows that synthesizing and capping GNPs with various amino acids not only alters the size and geometry, but also the capability of AAGNPs as parts of recognition elements of sensor systems. This study highlights the potential of amino acid-mediated green synthesis as an environmentally friendly and versatile approach for developing functional nanomaterials with tunable sensing capabilities for pollutant detection.Conference Object Numerical Investigation of Ventilation Design Effects on Exposure Equality in Ventilated Classrooms(Springer International Publishing AG, 2026) Yetis, Ahmetcan; Sofuoglu, Sait Cemil; Sofuoglu, Aysun; Cobanoglu, Nur; Sahin, Cagri; Karadeniz, Ziya HaktanSchools are crucial buildings, where indoor air quality (IAQ) impacts the academic performance, productivity, and health of students and teachers. Therefore, implementing efficient ventilation systems in high-occupancy areas, particularly classrooms, is essential to prevent accumulation of polluted air and to provide exposure equality among children. This study aims to examine the impact of ventilation design on the spatial pollutant exposure equality in classrooms, specifically focusing on the distribution of submicron particulate matter (PM1) and the air change efficiency. Anthropological data was utilized to accurately represent the geometry of student mannequins, and a three-dimensional classroom model was created accordingly. Steady numerical analyses were conducted parametrically to calculate the air change efficiency for different air flow rates, diffuser angles (diffuser angle: 22 degrees horizontally and 60 degrees and 30 degrees vertically; air flow rate: 3.7 and 8 l/s/person), and PM injection. The three-dimensional airflow and pathogen particle movement in the classroom were determined using realizable k-epsilon turbulence modelling and discrete phase modelling (DPM). For exposure equality, the case with a flow rate of 3.7 l/s/person and a diffuser angle of 60 degrees is the worst design. Front-row students consistently face higher velocity and PM1 concentrations compared to others in all cases, but increasing the flow rate to 8 l/s/person and reducing the diffuser angle to 30 degrees improved the concentration distribution by up to 75%. At a flow rate of 8l/s/person, the decrease in the diffuser angle resulted in a decrease of up to 86% in the velocity values around the heads of these students.Article Navigating the Paradox of Consensus: Insights from Amsterdam and Ankara(SAGE Publications Inc, 2026) Ozdemir-Ulutas, Esin; Tasan-Kok, TunaThe critique of consensus within post-political literature has characterised the relationship between consensus and conflict as oppositional. Drawing on an in-depth re-reading of Habermasian theory of communicative action and a comparative inquiry in to the experience of planning practitioners in Amsterdam and Ankara informed by Forester's critical pragmatism, we propose that the relationship between consensus and conflict is dynamic and mutually constitutive, rather than simply oppositional, with planners serving as expert authorities. In both cities, consensus is regarded as an alternative to authoritarian approaches in planning and as a mechanism for addressing citizens' demands and disagreements. The findings further indicate that planners in Amsterdam actively pursue consensus-seeking, while those in Ankara encounter difficulties due to political pressure. Although there are challenges and limitations associated with this approach, the article concludes that discontinuing the use of consensus is not considered a viable option in local planning processes.Article Investigation of Viscerotropism-Associated Genes in Leishmania Tropica Strains Causing Visceral Leishmaniasis(Frontiers Media SA, 2026) Gündüz, Cumhur; Cavus, Ibrahim; Dinc, Melike; Vardarlı, Aslı Tetik; Tunali, Varol; Beyaz, Merve; Ozbilgin, AhmetLeishmania infantum is the predominant etiological agent of visceral leishmaniasis (VL), whereas Leishmania tropica is classically associated with cutaneous leishmaniasis (CL). However, sporadic reports from Turkiye and other endemic settings indicate that L. tropica may also cause VL, suggesting that specific parasite determinants enable adaptation to visceral tissues. The molecular basis of this phenotypic shift remains insufficiently understood. This study aimed to define genetic factors associated with visceralization in L. tropica by integrating clinical isolates from patients with VL and CL and evaluating candidate genes involved in oxidative stress defense, mitochondrial function, proteolysis, and metabolic adaptation. Fourteen patients diagnosed between 2012 and 2022 were included, comprising seven VL and seven CL cases. Parasites isolated from clinical specimens were cultured and genotyped as L. tropica using real-time quantitative PCR targeting the internal transcribed spacer 1 region. Gene-expression profiling by qRT-PCR focused on Cytochrome C Oxidase subunit IV, Metallo-peptidase (Clan MA(E), Family M32), Oligopeptidase B, Peroxiredoxin 1, Peroxiredoxin 2, Pyruvate kinase, and Succinyl-CoA:3-ketoacid-coenzyme A transferase. Compared with CL isolates and reference strains, VL isolates demonstrated markedly increased mRNA expression of Peroxiredoxin 1 and Peroxiredoxin 2 and Cytochrome C Oxidase subunit IV, with approximately 17-fold and 21-fold elevations, respectively. Additional increases were observed in metallo-peptidase, Oligopeptidase B, and Succinyl-CoA:3-ketoacid-coenzyme A transferase, supporting a broader program of antioxidant, mitochondrial, proteolytic, and metabolic adaptation. Targeted next-generation sequencing identified multiple coding variants in Oligopeptidase B and the M32 metallo-peptidase, suggesting potential contributions to phenotypic divergence between VL and CL isolates. Viscerotropic L. tropica exhibits a distinct molecular profile characterized by enhanced antioxidant defense, mitochondrial activity, proteolytic capacity, and metabolic flexibility. These findings identify plausible genetic drivers of visceralization, expand current understanding of L. tropica pathogenicity, and support the development of testable diagnostic and therapeutic targets. Collectively, the data challenge the prevailing view of L. tropica as an exclusively cutaneous parasite. The results also provide a rational framework for future functional validation studies aimed at clarifying causality, refining biomarkers, and prioritizing parasite specific intervention strategies in endemic settings globally.