National Institute of Applied Science and Technology

Abstract We are grateful for Darwish's interest in our paper, Fouad et al. (2025, https://doi.org/10.1029/2024EF004883 ). In this reply, we show that Fouad et al. (2025, https://doi.org/10.1029/2024EF004883 ) did not attribute building collapses in Alexandria solely to hydroclimatic factors, as stated in the comment. Instead, we emphasize that hydroclimatic drivers are presented as accelerators, with other anthropogenic influences explicitly stated in the original paper. Moreover, our response proves that Darwish (2026, https://doi.org/10.1029/2025ef006885 )'s simplistic statistical approach is physically incorrect and obscures absolute risk by normalizing actual building collapse rates to the total number of buildings within a city. Furthermore, our reply shows that the comment conflates the distinct measurement of soil relaxation using shallow isotope mapping at the city scale, as conducted in Fouad et al. (2025, https://doi.org/10.1029/2024EF004883 ), with deep structural geotechnical assessments for foundation design of individual buildings. The utility and complementarity of both methods are already discussed in Fouad et al. (2025, https://doi.org/10.1029/2024EF004883 ). We acknowledge that the statement on the “7,000 at‐risk buildings” is only mentioned in the abstract and is inadvertently missed in the main text; however, the calculation leading to this result is detailed in our supplementary data set and methods. Accordingly, Darwish (2026, https://doi.org/10.1029/2025ef006885 )'s comment, while appreciated, misinterprets Fouad et al. (2025, https://doi.org/10.1029/2024EF004883 ) and overlooks the contemporary literature on Alexandria's hydrogeological and coastal dynamic contexts and their implications for infrastructure instability.

Trace metals are frequently detected in aquatic ecosystems, raising serious concerns due to theirtoxicity to aquatic organisms and their potential transfer to humans through the food chain. Consequently, there is an increasing demand for eco‐friendly bioremediation strategies that preserve aquatic biota and environmental integrity, unlike conventional chemical or physical treatments. Microalgae, particularly Chlorella sp., have emerged as promising candidates for metal removal, especially when immobilized within biocompatible matrices. This study evaluated the cadmium (Cd 2+ ) binding capacity of Chlorella sp. and investigated the kinetics of cadmium retention using alginate‐immobilized cells. For the first time, potentiometric sensors incorporating alginate‐immobilized Chlorella sp. were developed and applied for cadmium detection. The system exhibited a reproducible near‐Nernstian response to Cd 2+ , with a slope of 29 mV per decade and a linear detection range from 10 −3, 42 to 10 −1, 67 M. The maximum cadmium fixation rate reached 205 μM Cd g −1 Chlorella sp. min −1 , with complete saturation achieved within 35–40 min. Cadmium retention involved coupled physicochemical interactions and biologically mediated processes, progressing through four successive stages: rapid surface binding, partial metal release, a transient adjustment phase, and establishment of a stable dynamic equilibrium. These findings highlight alginate‐immobilized Chlorella sp. as a sustainable efficient tool for cadmium bioremediation.

Tunisian traditional fermented foods represent a valuable cultural heritage transmitted across generations and are highly appreciated by consumers for their distinctive flavours, textures, and nutraceutical value. This review provides the first comprehensive and exclusive overview of lactic acid bacteria (LAB) associated with Tunisian traditional fermented foods, both plant- and animal-based. The overview integrates data across dairy, meat, fish, vegetable, and cereal matrices, highlighting the central role that LAB play in the processing of these foods, driving fermentation and shaping the quality and safety of final products. During fermentation, LAB produce a variety of bioactive metabolites, including organic acids, antimicrobial compounds, exopolysaccharides, enzymes, and vitamins, which enhance food safety, shelf life, nutritional quality, and health-promoting potential. The studies include evidence of LAB's long history of safe use by humans, including the characterisation of autochthonous strains with protechnological, bioprotective, and probiotic properties, providing candidates for the design of starter, protective and probiotic cultures. By consolidating evidence on the relevance of microbial diversity, this review positions Tunisian LAB as valuable resources for both traditional food valorisation and innovative food system development. Importantly, key knowledge gaps are identified, including the limited application of omics-based tools, insufficient genomic safety assessments, and the lack of systematic analysis linking LAB diversity with the desired attributes to promote innovations. Overall, this review provides a structured framework for the valorisation of Tunisian agrofood heritage, bridging artisanal knowledge with modern food microbiology and offering strategic directions for future research, industrial translation, and sustainable innovation in fermented foods.

Abstract Erlotinib resistance remains a critical barrier in treating EGFR-mutant non-small cell lung cancer (NSCLC). While distinct resistance mechanisms have been identified, the temporal evolution of transcriptional states and the role of non-coding RNAs in this process remain poorly understood. To address this, we performed a secondary single-cell RNA sequencing (scRNA-seq) analysis of PC9 cells treated with Erlotinib (GEO Accession: GSE149383). We employed pseudotime trajectory inference (Monocle3) and rigorous in silico modeling to map resistance evolution and predict miRNA-lncRNA interactions. Our trajectory analysis revealed a biphasic evolution of resistance: an early phase characterized by ribosomal stress responses (RPS5, RPL21) followed by a late proliferative phase driven by cell cycle regulators (CENPF, HMGB2). Notably, the long non-coding RNA NEAT1 showed dynamic upregulation during this transition. Computational modeling identified miR-124-3p as a high-confidence regulator of NEAT1, with structural analysis confirming a thermodynamically stable interaction (Δ G = − 14.8 kcal/mol). These findings suggest that Erlotinib resistance is not a static state but a dynamic process involving sequential transcriptional reprogramming. We propose the miR-124-3p/NEAT1 axis as a potential therapeutic target to disrupt the stress-adaptation phase of drug resistance.

The antioxidant, antibacterial, and repellent activities of essential oils from different plant parts of Tunisian Schinus molle were determined. Gas chromatography-mass spectrometry (GC-MS) analyses revealed that the stems' essential oils are dominated by Geraniol (20.94%), α-eudesmol (28.42%), and Germacrene B (13.6%). Leaves and seeds were characterised by α-limonene (30.43 and 37.66%, respectively) and α-eudesmol (12.48 and 8.55%, respectively) as the most abundant compounds. The antioxidant activity of essential oils was assessed using two techniques: inhibition of the free radical DPPH and the iron reduction effect (FRAP). Considerable levels of antioxidant activities of the investigated essential oils were highlighted. Stems essential oils revealed the best activity with 0.42 mg/mL for the antiradical capacity and 35.52 µmoL TROLOX/g HE for the FRAP assay. Based on the determination of the diameter of inhibition, the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC), a high antibacterial activity was observed against Gram-positive strains Staphylococcus aureus and Enterococcus feacium (57 and 53 mm, respectively) in the presence of stems' essential oils. For repellent activity, the essential oils of stems showed the best potential, followed by the oils of seeds. The differences in the biological activities of the tested essential oils were attributed to their essential oil compositions. Consequently, S. molle stems essential oil can serve as a natural source of bioactive compounds for use in the food and pharmaceutical industries.

Abstract In breast cancer diagnosis, the accurate and efficient analysis of ultrasound images is essential, particularly for distinguishing benign from malignant tumors. Ultrasound imaging is a widely used, non-invasive diagnostic tool that provides valuable insights into tumor characteristics. This study presents a deep learning approach for breast cancer classification using transfer learning (TL) models across three distinct datasets: BUSI, BUSI-UCLM, and GDPH&SYSUCC. A significant challenge addressed was the inherent class imbalance within these datasets. To mitigate this, a comprehensive evaluation of various data sampling techniques was conducted, with Random Over Sampling (ROS) emerging as the most effective method for balancing the data. Among multiple TL models assessed, the ResNet50V2 architecture consistently demonstrated superior performance across all metrics. When trained and validated on the individual 1 datasets, the ResNet50V2 model achieved an accuracy of 95.44%, an F1-score of 95.25%, and an AUC of 99.21% on the BUSI dataset; 97.22% accuracy, 97.32% F1-score, and 99.90% AUC on the BUSI-UCLM dataset; and 95.72% accuracy, 95.72% F1-score, and 98.37% AUC on the GDPH&SYSUCC dataset. Following this individual evaluation, a combined dataset was created, which consisted of 3971 images distributed across benign (1497), malignant (1819), and normal (247) classes. On this combined, more challenging dataset, the ROS-enhanced ResNet50V2 model maintained its strong performance, achieving a final accuracy of 93.96%, an F1-score of 93.99%, and an AUC of 98.70%. These results highlight the efficacy of using ROS to address class imbalance and the robustness of ResNet50V2 as a transfer learning backbone for breast cancer classification across heterogeneous ultrasound datasets. The findings underscore the potential of this approach to enhance diagnostic accuracy and support clinical decision-making.