Gastric cancer remains one of the leading causes of cancer morbidity and death worldwide. Despite advances, current treatments remain costly, have low success rates, and cause adverse effects, prompting research into plant-based alternatives as a potential solution. Crataegus azarolus contains compounds with health potential and minimal side effects. This study investigates the potential antiproliferative, migratory, and apoptotic effects of C. azarolus methanol and acetone extracts of fruits and leaves on human gastric adenocarcinoma cell lines (AGS) and human fibroblast cell lines. AGS cells were treated with different concentrations of methanol and acetone fruits (MF and AF) extracts (10, 50, 100, 250, and 500 μg/mL). The results demonstrates that MF extract showes significantly higher anti-proliferative activity at 500μg/mL concentrations, when estimated by 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay (p < 0.001). The calculated half-maximal inhibitory concentration value is 300μg/mL. Furthermore, MF extract showes a significant (p ˂ 0.01) cellular migration inhibitory effect. The clonogenicity of AGS is considerably inhibited (p < 0.001), resulting in a decrease in holoclone production in AF and acetone leave (AL) extracts. Flow cytometry reveales induction of apoptosis, with highest 31.4% MF. Early growth response-1 expression increases dramatically (p < 0.001) in AF extract, while enhancer of zeste homolog-2 expression increases significantly (p< 0.001) when treated with MF and AL. Phosphatase and tensin homolog and N-myc downstream regulated gene-1 are both highly upregulated in MF and AF extracts, respectively. In conclusion, C. azarolus demonstrates anti-cancer effects on AGS cell line and may have potential as a source of anticancer compounds.
Conventional two-dimensional (2D) mammography compresses complex breast anatomy into a single projection, obscuring lesions through tissue overlap and limiting reliable tumor sizing. Although tomosynthesis and computed tomography recover depth information, these modalities are not universally available and add dose, time, and cost. We propose a single-view 2D-to-three-dimensional (3D) reconstruction framework that couples a 3D U-Net with a Generative Adversarial Network (GAN) to synthesize anatomically coherent 3D mammograms from a single 2D input. The end-to-end pipeline performs (i) seed stack formation and normalization, (ii) adversarial 3D reconstruction, and (iii) lightweight post-processing—Gaussian smoothing, global thresholding, and 26-connected component analysis—to localize regions of interest (ROI) and compute voxel-accurate lesion volumes in mm³/cm³ from pixel and slice spacings. Training combines weighted GAN and reconstruction losses to balance perceptual realism with structural fidelity. On a held-out test set, the method achieves a mean squared error of 0.0059, a mean peak signal-to-noise ratio of 24.13 dB, a mean absolute error of 0.0283, and a structural similarity index of 0.9296, outperforming interpolation and non-adversarial 3D U-Net baselines. Qualitative renderings show preserved parenchymal textures, smooth interslice transitions, and precise ROI overlays, while a red isosurface visualization highlights the 3D lesion extent for volumetry. The approach is simple to train, reproducible, and compatible with standard imaging toolchains, delivering clinically actionable volumetric measurements without multiview acquisition. Future work includes multiview supervision (craniocaudal and mediolateral oblique) to strengthen depth consistency, uncertainty maps to qualify reconstructions, and learned ROI segmentation with topology-aware objectives, advancing low-overhead 3D support in routine mammography.
Lung cancer is a leading cause of cancer mortality worldwide. Steroid hormones and complement proteins may contribute to lung cancer progression and immune regulation. This case-control study enrolled 111 individuals, including 66 newly diagnosed patients with lung cancer and 45 healthy controls. Blood samples were collected from all participants between 7:00 and 9:00 AM following an overnight fast of at least 8 h before the initiation of any treatment in the lung cancer patient group. Serum steroid hormones were statistically higher in the lung cancer group than in the control group (p = 0.0001). In addition, the levels of serum complement C3 (p = 0.006) and C4 (p < 0.0001) in the lung cancer group were higher than those in the control group. Levels increased with disease progression. Elevated serum concentrations of pregnenolone (odds ratio [OR] = 5.12, p = 0.036) and progesterone (OR = 7.20, p = 0.023) were significantly associated with an increased risk of lung cancer. The discriminatory power for serum cortisol (area under the curve [AUC] = 0.927), progesterone (AUC = 0.948), dehydroepiandrosterone (AUC = 0.935), and complement C4 was excellent (AUC = 0.962). Increased expression of several steroid hormones and complement proteins is linked to lung cancer and continuously increases with the development of the disease. Most of the steroid hormones and complement C4 showed a favorable capacity for early diagnosis and risk prediction of lung cancer. Nevertheless, pregnenolone and progesterone are now identified as the most relevant independent biomarkers.
For many low-resource languages, including the central Kurdish language (CKL), building effective natural language processing (NLP) tools has been a challenge. This is due to the lack of annotated text. Without a large corpus that specifies how words function grammatically, it is difficult to perform basic tasks such as part-of-speech (POS) tagging, which is the building block of many language technologies. To address this issue, this study presents the first comprehensive POS-tagged corpus for CKL. This dataset consists of 108,680 words manually tagged with 86 tags. Unlike simpler tagging schemes, the 86 tags account for the complexity of Kurdish grammar and allow a single word to have multiple valid tags, reflecting the language’s natural ambiguity. Using this resource, this study benchmarks a range of deep models, including neural networks such as bidirectional long short-term memory (BiLSTM). To address the ambiguity challenge, this paper introduces a new method, adaptive tag cycling within the BiLSTM that trains the model to consider all possible tags. The most advanced model in this study, an ensemble of neural sequence taggers, achieves 92.3% accuracy with stop-words retained and 89.5% with stop-words removed on broad grammatical categories (main tags). On the full fine-grained tagset (detailed tags), the same model attains 79.0% accuracy with stop-words and 76.2% without stop-words. Therefore, this study provides two key contributions: (i) a new dataset that supports future Kurdish NLP research, and (ii) a strong performance benchmark for CKL POS tagging.
Based on the available peer-reviewed literature, Kurdish handwritten recognition remains at a nascent stage, with existing models limited to isolated character, digit, and word recognition. The demand for digitizing Kurdish handwritten documents has grown rapidly, particularly in regions undergoing government digitization, such as the Kurdistan Regional Government. The primary obstacle hindering progress beyond isolated characters or words is the absence of text-based handwritten datasets. To address this gap, a comprehensive Kurdish handwritten dataset encompassing paragraph-, line-, and word-level samples is first introduced. A recognition architecture combining DenseNet-121 as the CNN backbone with a Transformer-based encoder-decoder for sequence modeling is then proposed for Kurdish handwritten line recognition, representing one of the earliest reported efforts in this domain. To augment the limited line-level training data, a constrained recipe-based synthetic line generation framework is developed that concatenates real handwritten word images while enforcing text uniqueness, single-writer consistency, and leakage-free data partitioning. Additional training strategies are investigated, including cross-lingual transfer learning from Arabic handwritten data and fixed-content handwritten line integration for improving recognition accuracy. The best configuration achieved a character error rate (CER) of 0.0593 and a word error rate (WER) of 0.3083 without language model integration, further reduced to a CER of 0.0534 and a WER of 0.2746 with an 8-g language model. The source code and trained models are publicly available at: https://huggingface.co/Karez/ KHLR.
Synthetic dyes found in wastewater create substantial environmental hazards and health risks, which require effective and sustainable wastewater treatment methods. The present study investigates using Cyperus rotundus as a natural adsorbent so that methyl orange is removed from aqueous solutions under a batch system. A total of 71 experimental runs were conducted in this study so that the impacts of varying dye concentration between (5 and 100) ppm, pH range between (3 and 11), a contact time increasing at discrete steps between (10 and 90 min), and finally a varying temperature from (25 to 45 degrees C) on the adsorption process were assessed. A 97% dye removal efficiency was obtained at a dye concentration of 10 ppm, pH of 4, and contact time of 90 min. Isotherm analysis revealed that the Langmuir model best fits the data with 2.25 mg/g adsorption capacity and a coefficient of determination (R-2 = 0.657). The kinetic data followed a pseudo-second-order model, indicating that chemisorption is the governing process (R2 > 0.95). Thermodynamic analysis proved that this adsorption is both spontaneous (with a standard Gibbs free energy change ranging from OG degrees =-4.76 to-5.69 kJ/mol) and endothermic (enthalpy change OH degrees = 9.10 kJ/mol). Both scanning electron microscopy and Fourier transform infrared spectroscopy confirmed the presence of dye adsorbent interactions when both structural and functional changes of the groups that occurred after adsorption were tested. The results show that C. rotundus is a low-cost biosorbent in methyl orange removal and can serve as a sustainable wastewater treatment material.
Breast cancer metastasis is one of the most consequential clinical problems in breast cancer, as it is responsible for most of the breast cancer attributable deaths in women globally. Timely detection of metastasis progression is crucial for improving therapies and enhancing patient survival. Although deep learning approaches, especially convolutional neural networks (CNNs), still fail to model global dependencies, they cannot work under weak supervision, and unable to generate interpretable predictions. Transformer-based models, on the other hand, provide a deeper contextual knowledge; but they are insensitive to local patterns and require large datasets. To overcome these issues, we present MetaBreastAI, an explainable dual-stream deep learning framework comprising a CNN branch fused with the convolutional block attention module and a Transformer stream. Each of these parallel branches captures spatial and contextual features, respectively, which are fused by a multi-instance learning approach for partially supervised classification. The proposed model has been tested on benchmark computed tomography scan dataset, it enables identifying one of the distinct features of MetaBreastAI and apply feature attribution methods to both visual and quantitative experiments, demonstrating that MetaBreastAI achieves better performance (macro-averaged F1-score: 91.7%; area under the curve: 96.1%) compared to each branch independently, and hybrid baseline models. By highlighting the lesion's location, the heatmap supports interpretability. Due to the explainable detection of lesions in metastatic cases, our hybrid model provides a scalable and clinically feasible strategy through overcoming the downsides of earlier research, then it improves the reliability of AI-assisted techniques in medical decision-making.
green synthesis of bis-thiazolidinone-thiadiazole hybrids (6a-h) exhibiting mild to potent antibacterial activity is reported. The hybrid compounds were synthesized in one-pot three-component Microwave-assisted reaction under solvent-free condition from the corresponding bis-thiadiazole amines (3) reaction with various aldehydes, and thioglycolic acid. The products were obtained in acceptable yields and characterized based on spectroscopic techniques (Fourier transform infrared, proton nuclear magnetic resonance, carbon-13 nuclear magnetic resonance, and carbon-attached proton test nuclear magnetic resonance). Antibacterial screening was performed on the synthesized compounds (6a-h) against both Grampositive strains (Staphylococcus and Streptococcus) and Gramnegative strains (Escherichia coli and Klebsiella pneumoniae), with their efficacy compared to the reference antibiotics ciprofloxacin and vancomycin. Overall, the synthesized compounds demonstrated highly antibacterial properties. Among the series, compounds (6b, 6d, and 6e) exhibited broad-spectrum and potent levels of antibacterial activity against species; furthermore, they showed notable activity against K. pneumoniae as the most antibacterial-resistant pathogen. The result of Molecular docking studies indicated that OH, OCH3 substituents, such as in compounds (6b), enhanced binding ability interaction and producing conventional hydrogen bond with the minimum energy affinity (triangle G).
Epilepsy is a neurological condition that is caused by an imbalance of excitation and inhibition in the brain. Many factors influence this imbalance, including enzymes involved in producing neurotransmitters, the balance of electrolytes, and the effect of trace elements. Thus, this study aimed to investigate the levels of glutamate dehydrogenase 2 (GDH2), NAD-dependent ADP-ribosyl transferase sirtuin 4 (SIRT4), electrolytes (Na, K, and Cl), and trace elements (Zn, Cu, and Se) in epileptic patients and compare them with control groups. The blood serum from 30 focal epilepsy (FE) patients, 30 generalized epilepsy (GE) patients, and 30 control groups was collected in Erbil City Hospital. The results of the study showed a significant decrease in GDH2 with a P < 0.0001 for the FE and GE groups. SIRT4 levels were significantly increased in both groups, compared to the control group. Furthermore, levels of Na, Zn, and Se were markedly reduced, while no significant variations were observed in K, Cl, and Cu levels between epileptic patients and control groups. The ROC curve and area under the curve (AUC) results for GDH2 (0.8092), SIRT4 (0.8956), sodium (0.8673), zinc (0.9711), and selenium (1). The results of the artificial neural networks showed that selenium, zinc, sodium, GDH2, and SIRT4 had the greatest impact on how the network classified patients as having epilepsy or being healthy. The results of GDH2, SIRT4, Na, Zn, and Se may be utilized as parameters for detecting the state of epileptic patients and used as beneficial biomarkers
Reinforced concrete (RC) columns are fundamental structural elements in buildings responsible for transferring loads from superstructures to the foundations. Improving their performance under different load conditions is important for ensuring structural safety and long-term durability, particularly to increase the flexural strength of columns. Because carbon fiber-reinforced polymer (CFRP) composites are so strong, lightweight, corrosionresistant, and simple to install, they should be used instead of traditional strengthening methods. This study examines the flexural strengthening of self-compacting RC L-shaped columns under axially and biaxially eccentric loads by using longitudinally applied CFRP. Six cases of columns in total were prepared and tested, including three unstrengthened columns and three flexurally strengthened with CFRP, which were tested at eccentricities (ex = ey ) of 0 mm (pure axial load), 50 mm, and 100 mm. The experimental findings showed that various loading conditions led to different failure processes. While concentrically loaded specimens failed suddenly as a result of brittle concrete crushing and reinforcement bar buckling, columns subjected to eccentric loading showed combined flexural-compression failure characterized by tension-side cracking, lateral displacement, and compression-side concrete crushing. There is less impact on increasing axial capacity at (ex = ey = 0), with an improvement of roughly 6%. The load capacity improved by up to 27% at (ex = ey = 50 mm) and 39% at (ex = ey = 100 mm) under eccentric loading. Results show loading situations have a major impact on the performance of CFRP, with flexural improving moment capacity, ductility, and fracture control under eccentric loads.
The low lubricity and high filtration loss of water-based drilling fluids (WBDFs) hinder the drilling industry and are the cause of high torque and drag, while also contributing to nuisance conditions such as wellbore instability. This study presents a new, environmentally friendly SiO2 nanoparticle (NP) made from rice husk ash (RHA), one of the byproducts of agriculture, as a multifunctional additive for WBDFs. The RHA-SiO2-NPs were produced by an alkaline extraction method and characterized using X-ray diffraction and scanning electron microscopy. RHA-SiO2 NPs were incorporated (0.1–1.0 wt%) into a bentonite-based mud and evaluated for rheology, API fluid loss, and lubricity. The results showed that the addition of only 0.5 weight % RHA-SiO2 resulted in a 60% reduction in fluid loss (16.8–6.8 mL) and a 39% improvement to the lubricity coefficient (0.28–0.17) in comparison to the base fluid. The presence of the RHA-SiO2-NPs also increased the yield point and gel strengths when looking at the rheological properties, indicating improved hole-cleaning capabilities. RHA-SiO2 NPs demonstrated multifunctional performance and potential for sustainable drilling applications.
The current study examines the single-nucleotide polymorphism of the GUCY2D gene in blind patients with inherited Leber congenital amaurosis (LCA) from a molecular, medical, and genetic perspective. The study involved 33 patients with blindness and 11 healthy controls in Erbil Province, Kurdistan Region, Iraq. LCA is the most severe phenotype of inherited retinal diseases, marked by genetic and clinical heterogeneity. The goal of this work was to confirm and identify the underlying mutations. The visual evoked potentials (VEPs) test, fundus photography, and ophthalmic examination were the foundation for clinical studies. Each participant’s genomic DNA was extracted to ascertain the frequency of mutations in the Kurdish nation. Primers were designed for the GUCY2D gene, encompassing exons 15, 16, and part of 17, with introns situated between those exons. Potentially pathogenic mutations were detected using Sanger sequencing analysis of the GUCY2D gene. Sequence data were analyzed to identify known, unknown, or novel mutations using FinchTV, BioEdit, and databases from the National Center for Biotechnology Information, including BLASTN, BLASTX, and ClinVar, along with Mutation Surveyor and MaxEntScan. Three novel splicing mutations have been found in the GUCY2D of blindness patients. These mutations are: c.3043+11C>T, c.3044-7G>T, and c.3043+20G>C. The patients have included nystagmus, abnormal VEPs, photophobia, hyperopia, and early infantile onset of vision loss. The three new mutations in the GUCY2D of Kurdish blind patients are associated with LCA disease. A family history of blindness or the presence of affected relatives may represent important risk factors for LCA, supporting its inherited genetic basis.
In this work, the SunSolve ray tracer is used to investigate the effects of various surface structures on silicon (Si) passivated emitter and rear cell (PERC) solar cells. A Si substrate with a thickness of 170 µm is used. The studied surface structures include front-side inverted pyramids, cones, and spherical caps. Flat Si is used as a reference. The performance of these structures is evaluated across the 200–1100 nm wavelength range under AM 1.5G solar spectrum illumination at normal incidence. The weighted average reflectance (RWAR) is calculated from the reflectance profile over the entire spectral range. Among the results, the PERC solar cell with a pyramid texture demonstrates a short-circuit current density (Jsc) of up to 40.55 mA/cm2 and a conversion efficiency (η) of 21.76%. This represents a significant performance improvement over the other structures, attributed to enhanced broadband light absorbance and increased device efficiency. This study thus provides a detailed analysis of how different front-surface structures affect the performance of Si photovoltaic cells.
In Iraq, combined cycle gas turbine (CCGT) systems are the only and main process for power generation (electricity). Due to its importance and significance, accurate simulation is crucial for evaluating the process performance under different operation conditions. In this study, DWSIM software, open-source simulation tool, was used for modeling and assessing CCGT performance during part-load circumstances. This software used for its reliability in capturing operational behavior. The methodology consists of simulating CCGT response under the variation of the power by integrating part-load performance equations into DWSIM software. Generally, two cases included in this work; the first case covered a DWSIM simulation during part-load operation using data from previously published works that used Aspen HYSYS. On the other hand, the second case included performing a DWISM simulation using real filed data. The results gained were validated by comparing it with Aspen HYSYS simulation. The comparison between DWSIM and HYSYS showed a 3.8% variation in power generation, a close match in fuel flow, and a maximum variation of 4.36% in gas turbine efficiency. The findings of this work confirm that the DWSIM software is flexible, free, and reliable simulation tool for evaluating CCGT power plant’s performance and operation. In addition, this work presented that this software is suitable for use in both industrial and academic applications due to its performance compared to that of commercial simulators.
Breast cancer metastasis is one of the most consequential clinical problems in breast cancer, as it is responsible for most of the breast cancer attributable deaths in women globally. Timely detection of metastasis progression is crucial for improving therapies and enhancing patient survival. Although deep learning approaches, especially convolutional neural networks (CNNs), still fail to model global dependencies, they cannot work under weak supervision, and unable to generate interpretable predictions. Transformer-based models, on the other hand, provide a deeper contextual knowledge; but they are insensitive to local patterns and require large datasets. To overcome these issues, we present MetaBreastAI, an explainable dual-stream deep learning framework comprising a CNN branch fused with the convolutional block attention module and a Transformer stream. Each of these parallel branches captures spatial and contextual features, respectively, which are fused by a multi-instance learning approach for partially supervised classification. The proposed model has been tested on benchmark computed tomography scan dataset, it enables identifying one of the distinct features of MetaBreastAI and apply feature attribution methods to both visual and quantitative experiments, demonstrating that MetaBreastAI achieves better performance (macro-averaged F1-score: 91.7%; area under the curve: 96.1%) compared to each branch independently, and hybrid baseline models. By highlighting the lesion’s location, the heatmap supports interpretability. Due to the explainable detection of lesions in metastatic cases, our hybrid model provides a scalable and clinically feasible strategy through overcoming the downsides of earlier research, then it improves the reliability of AI-assisted techniques in medical decision-making.
the artificial intelligence (AI) supported by software-defined networking and network function virtualization has a significant impact on the performance, flexibility, and efficiency in the development of 6G network slicing. This article compares AI-driven 6G slicing networks with traditional rulesbased 5G networks, focusing on latency, data throughput, jitter, power efficiency, and bandwidth. NS-3 and MATLAB have been utilized to evaluate the networks performance. The comparison results show that AI-driven network slicing reduces average latency by 50%, boosts data throughput by 40-90%, reduces jitter by 50%, and improves power efficiency by 20-28% compared to 5G networks. These results indicate that AI-powered network slicing in 6G networks outperforms traditional methods, enabling trendier network management. This sets a standard for network segmentation research in the future deployment of 6G networks.
Duhok, a city in Kurdistan Region of Iraq, has important historical and cultural evident in its archaeological sites. The city center contains many neglected heritage buildings that necessitate documenting as a crucial preliminary step for their preservation, facilitating a comprehensive understanding of their architectural, historical, and social significance. Consequently, the study aims to document buildings in the study region employing several documentation methods and looking at various aspects, such as architectural, historical, social, and cultural dimensions. The second goal of the study is to build an automated model utilizing advanced machine learning techniques, such as convolutional neural networks, transfer learning, and neural architectural search, to create a robust model for identifying and classifying architectural styles across various regions of the Kurdistan Region. The initials result of the study show the unique attributes of Islamic, vernacular, modern, and postmodern architectural styles within Duhok's legacy. The machine learning categorization of the model is very accurate, highlighting its potential as a reliable analytical tool for identifying and classifying architectural styles in Duhok city and across Kurdistan Region.
The lack of security measures in Internet of Things (IoT) systems has made these tiny devices vulnerable to increasingly advanced and evolving cyber-attacks. Attack detection and prevention are one of the most promising approaches to mitigating these attacks. However, these techniques require more computing, memory, and energy than typical IoT devices can provide. In addition, the IoT network is distributed, heterogeneous, and dynamic. These limitations motivate this review to examine the effective use of machine learning and deep learning in detecting attacks on IoT systems. Despite the presence of prior studies on reviewing these techniques, there is still a gap in analyzing attack vectors and assessing the effectiveness of current detection techniques for IoT networks and environments, especially in terms of lightweight and real-time evaluation. In this work, a multidimensional taxonomy of IoT attacks and existing detection techniques is given. The included studies were critically analyzed to evaluate and assess their effectiveness, considering performance metrics, IoT system architecture, datasets, and deployment strategies. The core methodologies of the analyzed studies were examined to guide academia and industry in improving detection techniques. Results showed that most of the proposed techniques in the literature did not address IoT-specific requirements. However, techniques featuring lightweight, real-time, federated, and scalable solutions have been proposed, yet their practical effectiveness remains unvalidated. This review addresses key research gaps and future challenges, emphasizing the need for resource-efficient, adaptable detection methods that align with IoT constraints.
Synthetic dyes found in wastewater create substantial environmental hazards and health risks, which require effective and sustainable wastewater treatment methods. This study investigates the use of Cyperus Rotundus as a natural adsorbent to remove methyl orange in aqueous solutions under a batch system. A total of 71 experimental runs were conducted to examine the effects of initial dye concentration (5–100 ppm), pH (3–11), contact time (10–90 min), and temperature (25–45°C) on the adsorption process. The maximum removal efficiency of 97% was obtained at a dye concentration of 10 ppm, pH of 4, and contact time of 90. Isotherm analysis revealed that the Langmuir model can provide the best fit with a maximum adsorption capacity of 2.25 mg/g, and a coefficient of determination (R2 = 0.657) while kinetic data followed the pseudo-second-order model indicating that chemisorption is the dominant mechanism (R2 > 0.95). Thermodynamic analysis proved that this adsorption is both spontaneous (with a standard Gibbs free energy change ranging from ΔGo = - 4.76 to -5.69 kJ/mol) and endothermic (enthalpy change ΔH° = 9.10 kJ/mol). Scanning Electron Microscopy (SEM) and Fourier Transform Infrared Spectroscopy (FTIR) were used to confirm the presence of dye adsorbent interactions by testing structural and functional changes of the groups that occurred after adsorption. The results show that Cyperus Rotundus is a low-cost biosorbent in methyl orange removal and can serve as a sustainable wastewater treatment material.
In this study, the occurrence of two introduced terrestrial gastropod species: Monacha cartusiana (O.F. Müller, 1774) and Cochlicella barbara (Linnaeus, 1758), in the Kurdistan Region of Iraq, was investigated. Species identifications were made based on snail morphology. Snail shells were studied to determine their chemical composition and its effect on their success and establishment in the study area. Their organic components (protein and chitin) and inorganic matrix (elements) were analyzed and their correlations were investigated. Elemental analysis was performed using inductively coupled plasma - mass spectrometry and energy dispersive X-ray fluorescence for both species and their surrounding soil. Gravimetric analysis revealed that the main shell component was calcium carbonate (CaCO3) in the form of aragonite, comprising over 96% of the shells’ mass. Proteins and carbohydrates accounted for <4% of the shells. Fourier-transform infrared spectroscopy spectral analysis confirmed the presence of chitin in its β-allomorph form. Thirty-eight elements were detected. Of which, potassium (K), strontium (Sr), and rhenium (Re) were significantly different between the two species. A highly significant correlation was found between protein and carbohydrates, and CaCO3 indicating the role of the organic matrix in the formation of the mineral component. The survival of the studied species is partly attributed to a considerable overlap in their organic and inorganic chemical characteristics, along with notable differences.
