American University in Cairo

Abstract Background Atherosclerosis poses a significant global health burden, urging the need to discover new therapeutic targets. Emerging evidence suggests associations between gut microbiota and direct vascular bacterial infection in atherosclerosis. However, antibiotics failed to treat atherosclerosis, underscoring the complexity of the microbial community that promotes atherogenesis1. The involvement of fungi in plaque development remains largely understudied, with limited research using low-throughput methods. Purpose In this study, we aimed to investigate the potential of fungi as contributors to coronary artery infection–mediated atherogenesis by applying whole-metagenome analysis of aspirated thrombi from patients with atherosclerosis. Methods This study recruited nine patients with ST-segment–elevation myocardial infarction (STEMI) and one patient with non-STEMI. DNA was extracted from five non-host-depleted samples with prior mechanical lysis in liquid nitrogen and four host-depleted samples with prior collagenase treatment, followed by whole-genome amplification. In addition, one sample was extracted using both methods. Eight libraries passed QC and were sequenced using the Illumina NovaSeq platform. Sequencing reads underwent quality trimming and stringent host read removal. Fungal hits were identified via Bowtie2 mapping to the fungal RefSeq database, validated by BLAST and Kraken2. Species-level identification was performed by reference genome alignment. Results A total of twenty-one fungal genera were identified in six samples, whereas two samples showed no detectable fungi. Alternaria was the most prevalent genus detected in five samples (62.5%) and exhibited the highest relative abundance in two of them. The second most prevalent genus was Aspergillus, detected in four samples (50%). It displayed high relative abundance in two of these samples and co-occurred with Candida. Additionally, Malassezia was detected in three samples (37.5%). Conclusions These genera, except for Candida, were first identified in atherosclerotic thrombi in our study. Furthermore, compositional alterations of these genera, along with Exophiala, have been previously documented in the gut of patients with atherosclerosis, indicating a possible link between gut dysbiosis and direct vessel infection in atherosclerosis. Reinforcing this connection, Candida has been shown to evade monocytes, disseminate to distant organs through the “Trojan horse” mechanism, which is also associated with gut dysbiosis, and induce foam cell formation in vitro. An additional potential pathway involves fungal translocation as a sequela of long COVID syndrome. Our findings suggest a possible fungal gut–vessel axis in infection-driven atherosclerosis, involving fungus-laden monocytes and post-COVID-19 fungal translocation. This pathway could represent a novel therapeutic target, warranting further investigation.

The photophysical properties of a recently synthesized photoCORM, the Mn(I) tricarbonyl complex fac-[MnBr(CO)3(AQ)] (AQ = 8-aminoquinoline), with promising cytotoxic activity against triple-negative breast cancer, have been investigated by means of density functional theory (DFT) and time-dependent DFT calculations. Simulations in various solvents (water, DMSO, THF) confirmed the Mn-CO bonds follow a Dewar-Chatt-Duncanson model, only slightly modulated by solvent polarity. The optimized computational method accurately reproduced the UV-vis spectrum, identifying the lowest energy excitations as metal-to-ligand charge-transfer (MLCT) states with minor ligand-to-ligand charge-transfer (LLCT) contributions, whose extent depends on solvent polarity. Exploration of the triplet excited-state manifold revealed several low-lying metal-centered (3MC) states accessible from the initial singlet. High spin-orbit coupling and rapid intersystem crossing rates indicate that CO photorelease occurs through efficient population of these dissociative 3MC states, especially in polar media. These findings provide mechanistic insight into the photoactivation pathway of Mn(I)-based photoCORMs and establish a robust computational framework for designing efficient CO-releasing therapeutic agents.

In the original publication [...]

Abstract In recent years, a new form of Arabic has emerged to facilitate communication between younger generations, particularly with the advent of social media platforms. Globalization was one of the primary factors that increased the importance of the English language, particularly with the widespread adoption of technology and the dominance of various technological devices and platforms \cite{HAL}. This new form of Arabic is 'Arabizi, a portmanteau of Araby-Englizi, meaning Arabic-English, is a digital trend in texting Non-Standard Arabic using Latin script \cite{taha01}. The intensified use of Arabizi has given rise to a plethora of new research concerns about how to interpret this type of language using various machine learning approaches. Consequently, Natural Language Processing (NLP) might aid in deriving substantial insights, allowing sentiment analysis. This study will review the various approaches presented in the literature to address this topic. Then we tested a set of machine learning models, deep learning models, and tested out ensembles made with both. The results were that a fine-tuned Support Vector Machine (SVC) produced the best results with an accuracy of 0.63 and F1 score of 0.59.

Existing buildings consume a large portion of the total current energy production, especially in developing countries such as Egypt. Increasing energy demand, coupled with decreasing availability and increasing cost of conventional non-renewable energy resources, have encouraged a “building green” retrofit trend in order to maximize the energy performance of the built environment. This paper outlines the development of an Energy Retrofit Decision-Support System (ERDSS) for hot, arid climates that models building retrofit scenarios and determines the impact of each retrofit measure on the overall energy consumption of a proposed building retrofit program. The methodology combines building an energy simulation with a database-driven, budget-constrained optimization framework based on the Savings-to-Investment Ratio (SIR) to evaluate and prioritize retrofit measures. In addition, ERDSS determines the impact of each retrofit measure on the overall energy consumption of a proposed building retrofit program, ranks the retrofit measures according to their Savings-to-Investment Ratio (SIR) and uses optimization to develop a suggested retrofit program for a given budget. ERDSS is applied on a case study of an education building in New Cairo, Egypt, in order to illustrate the performance of the framework. Results show that savings for the commissioned retrofit, standard retrofits, and deep retrofits reached 15 percent, 35 percent, and 45 percent, respectively.

The motivation for this work is to study the cause and present mitigation for some challenges faced in preparing porous silicon. This enables benefiting from the appealing benefits of porous silicon that offers a wide range, simple technique for varying the refractive index. Such challenges include the refractive index values, sensitivity to oxidation, some fabrication parameters, and other factors. Additionally, highly doped p-type silicon is preferred to form porous silicon, but it causes high losses, which necessitates its detachment. We investigate some possible causes of refractive index change, especially after detaching the fabricated layers from the silicon substrate. Thereby, we could recommend simple but essential precautions during fabrication to avoid such a change. For example, the native oxide formed in the pores has a role in changing the porosity upon following some fabrication sequence. Oppositely, intrinsic stress doesn’t have a significant role. On another aspect, the effect of differing etching/break times on the filter’s responses has been studied, along with other subtle details that may affect the lateral and depth homogeneity, and thereby the process success. Solving such homogeneity issues allowed reaching thick layers not suffering from the gradient index. It is worth highlighting that several approaches have been reported; unlike these, our method doesn’t require sophisticated equipment that might not be available in every lab. To well characterize the thin films, it has been found essential that freestanding monolayers are used for this purpose. From which, the wavelength-dependent refractive index and absorption coefficient have been determined in the near infrared region (1000–2500 nm) for different fabricated conditions. Excellent fitting with the measured interference pattern has been achieved, indicating the accurate parameter extraction, even without any ellipsometry measurements. This also demonstrates the refractive index homogeneity of the fabricated layer, even with a large thickness of over 16 µm. Subsequently, multilayer structures have been fabricated and tested, showing the successful nano-manufacturing methodology.