Increasing resistance to insecticides and their accompanying off-target effects call for the development of biodegradable insecticides with selective metabolism-based targets. Anopheles gambiae trehalose-6-phosphate synthase ( Ag TPS), the first enzyme in the trehalose metabolic pathway, is essential for key developmental processes in the organism but absent in vertebrates, making it an attractive insecticidal target. This study aims to evaluate the inhibitory potential of neem-derived compounds against Ag TPS using in silico methods. The 3D structure was predicted, and its active-site residues were identified based on conserved TPS homologues. A library of 275 neem phytochemicals was compiled from IMPPAT and Dr Duke’s databases, alongside known substrates and inhibitors. Molecular docking was performed using AutoDock Vina at the glucose-6-phosphate (G6P), UDP-glucose (UDPG), and combined binding sites. Protein–ligand interactions were analysed using LigPlot+, while insecticide-likeness and ADMET properties were evaluated using QEPest and pkCSM. Several neem-derived compounds exhibited stronger binding affinities to Ag TPS than the natural substrates across all binding sites. Tannic acid showed a high binding affinity across all binding sites, with a range of -11.3 to -12.1 kcal/mol, showing interactions with conserved catalytic residues. Other compounds, including 1-cinnamoylmelianolone, nimbilin, 6-acetylnimbandiol, and 7-deacetyl-7-oxogedunin, also demonstrated strong and consistent binding, frequently outperforming known TPS inhibitors. ADMET analysis revealed that tannic acid possessed unfavourable pharmacokinetic and toxicity profiles. In contrast, limonoids such as 1-cinnamoylmelianolone and nimbilin displayed a more favourable balance between potency, insecticide-likeness, and predicted safety. Neem-derived phytochemicals show significant potential as inhibitors of Ag TPS, supporting Ag TPS as a promising target for selective mosquito control. While tannic acid demonstrates strong binding affinity, its poor drug-likeness limits practical application. The limonoids, such as 1-cinnamoylmelianolone and nimbilin, thus emerge as more viable lead candidates. These findings provide molecular insights into neem’s insecticidal activity and support further experimental validation for vector control applications.
The SCOP and CATH databases are publicly available resources that provide hierarchical annotations describing the structural and evolutionary relationships of protein domains. While considerable effort has been devoted to developing AI-based approaches for large-scale annotation within CATH, comparable developments for SCOPe remain relatively limited. CATHe and its successor CATHe2 have recently demonstrated the effectiveness of protein language model (pLM) embeddings combined with feed-forward neural networks for remote homology detection and superfamily classification in CATH. In this work, we adapt the CATHe2 framework to the SCOPe hierarchy and investigate its ability to perform protein family classification using amino acid sequences alone. Experiments were conducted on the ASTRAL SCOPe 2.08 dataset comprising 15,177 domains filtered at a maximum sequence identity of 40%. The proposed approach achieved F1-scores of 80.6% ±2.9% and 68.5% ±2.1% on the Top50 and Top1000 family classification tasks, respectively. When considering all 2,158 families containing at least two domains, an F1-score of 58.0% ±1.6% was obtained. These results demonstrate that pLM-based embeddings can effectively capture family-level relationships within SCOPe despite the challenging remote homology setting and the limited number of training examples available for many families, providing a promising step toward large-scale automated annotation of SCOPe domains.
Type 2 diabetes mellitus (T2DM) is a complex metabolic disorder involving insulin resistance, impaired glucose metabolism, mitochondrial dysfunction, and chronic inflammation. Imeglimin is a novel oral antidiabetic agent that improves mitochondrial function and insulin sensitivity, although its precise molecular mechanisms remain incompletely understood. This study employed an integrated network pharmacology approach combined with molecular docking, molecular dynamics (MD) simulation, and external validation to investigate the potential molecular targets and pathways of imeglimin in T2DM. Pharmacokinetic and drug-likeness properties were assessed using SwissADME, Molsoft, and pkCSM. Potential targets predicted by SwissTargetPrediction were intersected with T2DM-associated genes from GeneCards, followed by protein-protein interaction, Gene Ontology, and KEGG enrichment analyses. Molecular docking, redocking validation, 100 ns MD simulations, and MM-GBSA calculations were performed. External validation used DisGeNET and GEO datasets. Imeglimin exhibited favourable pharmacokinetic properties and Lipinski compliance. Forty overlapping targets were identified, with SLC6A3, HTR1A, GRIN2B, HTR2A, HTR2C, CA1, HTR1B, and EGFR as major hubs. Enrichment highlighted PI3K-Akt, dopaminergic synapse, calcium, and cAMP signalling. MD and MM-GBSA analyses supported stable interactions, particularly with SLC6A3, HTR2C, and GSK3B. These findings provide a systems-level basis for further experimental validation of imeglimin mechanisms in T2DM.
Objectives We aimed to design a multiepitope vaccine against hepatitis C virus (HCV) with the potential to elicit broad protective immunity against diverse HCV genotypes through computational design. Method : One hundred sequences of the most widespread subtypes of HCV were retrieved from the Los Alamos National Laboratory database, and consensus sequences of the NS5A and NS5B proteins were generated to construct an efficient vaccine for human use. Numerous B-cell lymphocyte, cytotoxic T-lymphocyte and helper T-lymphocyte epitopes located in the conserved domains of HCV proteins and similar in human and mouse alleles were evaluated criteria for use in the vaccine construct. Results Population coverage analysis identified that substantial parts of the worldwide population carry epitopes and their corresponding alleles within their genome, which confirms that the vaccine candidate is an effective agent against this lethal disease. Codon optimization and in silico cloning of eukaryotic and prokaryotic vectors validated the strong expression of the vaccine construct in the host systems. Molecular docking analysis revealed favorable binding affinities between the vaccine construct and toll-like receptors, which confirmed the vaccine’s capacity to induce robust and stable immune responses against the hepatitis C virus. Conclusion We incorporated new bioinformatic methods to design multiepitope vaccine against HCV. The development of a vaccine targeting multi genotypes of HCV could potentially provide for long-term protection against infection and may mitigate the risk of drug resistance across various genotypes and subtypes. Current research emphasizes the need for a vaccine that can elicit comprehensive immune responses, including both neutralizing antibodies and T-cell responses, to effectively address the genetic diversity of the virus. Future experimental studies are essential to validate predicted immunogenicity, structural stability, and protective efficacy before clinical consideration.
Alzheimer’s disease (AD) is a complex neurodegenerative condition marked by amyloid-β accumulation, tau pathology, and neuroinflammation, while current single-target therapies largely provide symptomatic relief without effectively halting disease progression. Therefore, the rational design and synthesis of novel small molecules capable of simultaneously modulating multiple AD-related targets represent a promising therapeutic strategy. Based on this rationale, novel pyrazoline derivatives were synthesized and their therapeutic potential was evaluated using an integrative computational framework. AD-related genes were retrieved from GeneCards, CTD, and DisGeNET databases, and hub genes were identified through topological and centrality analyses, followed by functional enrichment analyses to explore their biological relevance. Molecular docking was used to evaluate the binding affinities of the synthesized compounds toward the identified hub targets, while 3D-QSAR modeling was used to predict their inhibitory potential. A total of 63 common genes were identified across the three databases, with APOE, BACE1, and CSF1R emerging as key hub genes. Enrichment analysis revealed their involvement in PI3K–Akt, mTOR, NF-κB, TNF, AMPK, and FoxO signaling pathways associated with amyloid processing, tau pathology, and neuroinflammation. Among the evaluated compounds, compound 2a exhibited the strongest binding affinities toward the hub targets along with favorable drug-likeness, pharmacokinetics, toxicity, and promising IC 50 values (226–336 nM). The 3D-QSAR models showed high predictive reliability (R 2 = 0.93–0.96; Q 2 = 0.94–0.95). These findings highlight APOE, BACE1, and CSF1R as key therapeutic targets and identify compound 2a as a promising multi-target lead candidate for additional experimental validation in AD.
