Mofid University

Colon-targeted drug delivery as an oral drug delivery technique holds great promise for overcoming the limitations related to the modern medicine. These systems are highly able to protect the drugs from the low pH of the stomach and safely deliver them to the colon as they are pH-responsive. Poly(vinyl alcohol) (PVA) and poly(acrylic acid) (PAA) polymers have been widely used in drug delivery applications due to their excellent hydrophilicity, biocompatibility, and mechanical properties. Moreover, due to the presence of strong acidic functional groups in the PAA matrix, combination of PVA and PAA makes it possible to provide pH-sensitive drug delivery systems for colon-targeted drug delivery applications. Hence, pH-sensitive PVA-PAA-based hydrogels with various forms are suitable candidates for the safe delivery of different drugs such as non-steroidal anti-inflammatory drugs (NSAIDs), anticancer drugs, or even insulin to the colon. Therefore, the present review illustrates various forms of pH-responsive PVA-PAA-based hydrogels for colon-targeted drug delivery applications.

Therapeutic anti-diabetic peptides (ADPs) are an emerging class of clinically-relevant biologics with the potential to manage glucose (glycaemic) levels. However, the majority of existing computational pipelines are either narrowly focused or unverified. Here, we present a principled, end-to-end, and provenance-aware framework for the de novo generation, filtering and classification of ADPs. In terms of design, candidate peptides are derived from three complementary strategies: guided modification of functional motifs, recombination of conserved bioactive fragments, and a hybrid generative engine. Candidates are then subjected to biochemical triage (net charge, hydrophobicity, Boman index), homology screening, and APD-style predictors/calculators. For classification, we use a blend of (1) interpretable biochemical descriptors (net charge, hydrophobicity, Boman index), and (2) sequence-derived representations learned by a CNN-with-attention backbone, to parse local motifs and longer-range context. Classifier heads are automatically tuned with an Optimized Tree-structured Parzen Estimator (OptimizedTPE). Training used 238 experimentally validated ADPs with homology-aware splits (Train/Val/Internal-Test positives: 167/24/47) and a curated negative pool at a 2:1 ratio; additionally, Train-only weak-label augmentation added 412 screened positives (and matched negatives) for robustness. We report an evaluation on an independent, external panel of 180 peptides, fully disjoint from the training data in both source and time. On this unseen set, the model achieves ≈ 98.75% accuracy (F1 ≈ 0.985, precision ≈ 0.99, recall ≈ 0.98, specificity ≈ 0.99, ROC AUC ≈ 0.99). This suggests high sensitivity to true ADPs while tightly controlling false positives in the setting of realistic class imbalance. Taken together, these results make the framework a candidate for a reproducible, biologically-grounded, in silico screening layer for metabolic peptide therapeutics.

In the era of intelligent fitness technologies, the need for personalized training solutions that adapt to individual users’ physiological states, performance patterns, and exercise goals has grown considerably. This trend highlights the importance of advanced fitness recommendation systems capable of providing precise, adaptive, and data-driven training guidance. However, achieving such personalization requires continuous processing of sensitive physiological and behavioral data, which introduces critical security risks, including unauthorized access, data manipulation, and system vulnerabilities. To address these challenges, this paper presents a novel edge-enabled personalized fitness recommendation framework that integrates individualized training optimization with strong security guarantees. By leveraging the distributed and proximate processing capabilities of edge computing, the system minimizes communication exposure, reduces latency, and enables sensitive data to remain within secure edge environments or directly on the user’s device. This architectural design substantially mitigates security threats that commonly arise in centralized cloud-based processing. Comprehensive experimental evaluations demonstrate that the proposed framework not only delivers highly accurate and tailored fitness recommendations but also provides enhanced system robustness and data security compared with traditional cloud-centric approaches. The results indicate that edge computing offers a promising pathway toward secure, adaptive, and scalable personalized training technologies that support users’ performance improvement without compromising data integrity or system safety.

Wound healing is a complex process that can be compromised by bacterial infections, leading to delayed healing and an increased risk of complications. The aim of this study was to design and develop a novel antibacterial peptide, Healitide-GP1, which could be used in the treatment of infected wounds. A machine learning-based approach was used to identify the key features of WHPs and a genetic algorithm was used to generate new peptide sequences. Healitide-GP1 was synthesized and functionally validated. It showed high cytocompatibility (> 200 µg/mL in human dermal fibroblasts (HDF) and human keratinocytes (HaCaT)), improved wound closure (48% and 52% after 24 h, respectively) and strong antibacterial activity against Staphylococcus aureus (MIC: 12.5 µg/mL) and Escherichia coli (MIC: 25 µg/mL). Bioinformatics analyses revealed the unique hydrophobic motif and distinct evolutionary positioning of Healitide-GP1, suggesting a novel mechanism of action. These results emphasize the potential of Healitide-GP1 as a promising therapeutic candidate for the treatment of infected wounds. Our study demonstrates the successful integration of machine learning, bioinformatics and experimental validation in the development of therapeutic peptides and provides a valuable framework for the discovery of novel treatments for wound infections.

Targeted delivery of anticancer drugs holds great promise for enhancing therapeutic efficacy while minimizing adverse effects. The folate receptor (FR)-mediated approach offers a selective strategy to target cancer cells overexpressing FR. Bleomycin, an established antitumor antibiotic, suffers from limited efficacy due to poor diffusion into tumor cells. This study examined the anti-cancer potential of folate-targeted liposomal Bleomycin (FL-BLEOMYCIN) in comparison to non-targeted L-BLEOMYCIN on oral cavity cancer (CAL27). The study also investigated FL-Bleomycin's capacity to halt the cell cycle in the G2/M phase using flow cytometry. FL-Bleomycin was produced using thin-layer hydration, followed by incorporation of folic acid into nanoliposomes. To evaluate the release profile, drug release tests were carried out. Cytotoxicity of FL-Bleomycin, L-Bleomycin, and traditional Bleomycin was evaluated using cell viability assays. The cell cycle arrest caused by FL-Bleomycin was examined using flow cytometry. Finally, FL-Bleomycin uptake studies were performed to assess the internalization of FL-Bleomycin by CAL27 cells. Compared to L-Bleomycin and traditional Bleomycin, FL-Bleomycin showed noticeably more cytotoxicity against CAL 27 cells. The effective arrest of CAL 27 cells in the G2/M phase of the cell cycle by FL-Bleomycin was verified by flow cytometry. Uptake studies revealed increased internalization of FL-Bleomycin by CAL 27 cells compared to standard formulations. Drug release studies showed a consistent, non-explosive release profile. Cells treated with these nanoliposomes, compared to control groups, exhibited a dose-dependent decrease in the intensity of the 170-kDa EGF-R band as observed by Western blot analysis. The findings suggest that FL-Bleomycin is a potential method for delivering drugs precisely in tumors expressing folic acid receptors. Its potential for successful cancer treatment is shown by its higher internalization, improved cytotoxicity, and cell cycle prevention in CAL 27 cells. To find out how effective FL-Bleomycin is in vivo and whether it may be used to treat other FR-expressing tumors, more research is necessary.

Abstract Drug classification and target identification are crucial yet challenging steps in drug discovery. Existing methods often suffer from inefficiencies, overfitting, and limited scalability. Traditional approaches like support vector machines and XGBoost struggle to handle large, complex pharmaceutical datasets effectively. Deep learning models, while powerful, face challenges with interpretability, computational complexity, and generalization to unseen data. This study addresses these limitations by introducing a novel framework: optSAE+HSAPSO. This framework integrates a stacked autoencoder (SAE) for robust feature extraction with a hierarchically self-adaptive particle swarm optimization (HSAPSO) algorithm for adaptive parameter optimization. This combination delivers superior performance across various classification metrics. Experimental evaluations on datasets from DrugBank and Swiss-Prot demonstrate that optSAE+HSAPSO achieves a high accuracy of 95.52%. Notably, it exhibits significantly reduced computational complexity (0.010 seconds per sample) and exceptional stability (±0.003). Compared to state-of-the-art methods, the framework offers higher accuracy, faster convergence, and greater resilience to variability. Furthermore, ROC and convergence analyses confirm its robustness and generalization capability, maintaining consistent performance across both validation and unseen datasets. By leveraging advanced optimization techniques, the framework efficiently handles large feature sets and diverse pharmaceutical data, making it a scalable and adaptable solution for real-world drug discovery applications. However, the method's performance is dependent on the quality of the training data, and fine-tuning may be necessary for high-dimensional datasets. Despite these limitations, the optSAE+HSAPSO framework demonstrates transformative potential, significantly reducing computational overhead while improving classification accuracy and reliability. This work advances the field of pharmaceutical informatics by presenting a reliable and efficient framework for drug classification and target identification. These findings open promising avenues for future research, including extending the framework to other domains such as disease diagnostics or genetic data classification, ultimately accelerating the drug development process.

The demand for lightweight, high energy absorbing materials has increased, especially for high risk applications such as aerospace, transport and nuclear. This study aims at analyzing the design and behavior of biomimicking composite structures for structural applications based on the dermal armor system of pangolin under quasi-static loads. We consider two plant-based fibers, luffa and linen, as environmentally friendly substitutes for glass fibers. Using Taguchi, we quantitatively investigated the effects of five factors: fiber type, curvature radius, number of composite plies, trapezoidal scale Y1, and trapezoidal scale Y2 on TAE and SAE. Lateral compressive loads were applied to composite specimens prepared using unsaturated isophthalic polyester resin. It was observed that luffa fibers possess higher TAE and linen fibers possess higher SAE. The Taguchi method set the optimal values for the parameter levels for the prediction of TAE of 11.2431 J and SAE of 2.3677 J/g, while the actual experimental result brought the percentage of error to 5.76% for TAE and 3.94% for SAE. From the theoretical analysis of the RPP and Hollomon models, we gain insights into the energy dissipation process, such as curvature reduction and plastic hinge formation. It is hoped that this work will help to open up a wider field of bio-inspired designs for high-performance sustainable materials; in particular, the proposed framework provides clear ways of predicting the energy absorption characteristics of such structures.