MAPKAPK2 is a promising therapeutic target in numerous diseases. However, many MAPKAPK2 inhibitors are plagued by low solubility and permeability, and none have advanced through clinical trials. New computational methods utilizing deep learning can speed up inhibitor identification. This study aims to develop and validate a novel framework for MAPKAPK2 inhibitor discovery utilizing an ensemble of ten individual models trained on various feature sets. We trained DNN models using 21 molecular featurizers and 28 layer-size settings, and selected ten high-performing feature-architecture combinations to establish the ensemble. We explored various voting methods in conjunction with the ensemble and used the ensemble to generate a ranking of potential MAPKAPK2 inhibitors from an in-house compound set. Potential inhibitors satisfying Lipinski and Veber Rules not containing PAINS structures were selected for enzyme assay testing, and a molecular docking simulation was performed to investigate interactions. The individual model with the highest evaluation metrics was trained on functional-class fingerprints. Meanwhile, the ten-model voting ensemble reported an accuracy of 0.969 on a testing set. One novel MAPKAPK2 inhibitor, S021-0180, was identified out of seven tested with enzyme assays. The molecular docking simulation revealed critical ligand-residue interactions within the binding site. The novel computational framework was successful in identifying a novel MAPKAPK2 inhibitor as a promising inhibitor for further optimization in future studies. The established ensemble can be used to evaluate more compound sets for novel MAPKAPK2 inhibitors. Moreover, we anticipate that this new framework can be applied to all protein kinases for rapid compound screening.
The global outbreak and ongoing spread of the monkeypox virus (MPXV) have highlighted the urgent need for effective antiviral therapeutics. Here, we integrated artificial intelligence-based protein structure prediction, large-scale virtual screening, and experimental validation to identify preliminary hit compounds targeting MPXV. Using AlphaFold2, we predicted high-accuracy structures for seven essential MPXV proteins, including three polymerase-related and four surface proteins. Molecular docking of these targets against 6405 drugs from the ZINC15 world-approved subset generated a docking score dataset of 44,835 drug-protein pairs, from which numerous high-scoring compounds were identified. Focusing on A35R, we selected 26 compounds for experimental validation using surface plasmon resonance (SPR). Three compounds, including cepharanthine, eltrombopag, and simeprevir, exhibited measurable A35R‑associated binding signals with equilibrium dissociation constants (KD) in the micromolar range. Molecular dynamics (MD) simulations and molecular mechanics generalized Born surface area were employed for stability analysis and relative energetic assessment. Notably, all three have been previously reported to target other MPXV proteins, reinforcing their potential for repurposing. This work establishes AI‑driven structure prediction as a useful tool for identifying preliminary binding compounds. However, no antiviral activity has been demonstrated for these compounds; therefore, the three hit compounds warrant further optimization and biological evaluation. Our integrated approach provides a framework for rapid drug screening against emerging viral threats.
Environmental barrier coatings (EBCs) on SiC/SiC composites exhibit strong through-thickness heterogeneity arising from phase transitions, layered architectures, and microstructural defects, which can confound conventional per-layer statistical analysis. Here, we present a layer-aware data-driven framework that predicts hardness across the YbSi → mullite → Si → substrate stack using nanoindentation measurements and compositionally decoded layer descriptors. Individual indentation traces are concatenated into a continuous, normalized depth coordinate, enabling coating-scale hardness profiling without requiring perfectly segmented layers. To capture both local mechanics and depth-dependent coupling, we construct physics-aware synthetic descriptors that combine displacement, harmonic contact stiffness, normalized penetration, depth-coupling terms, phase-fraction features, and gradient-based (DIFF) measures. Redundancy in the expanded feature space is reduced via ANOVA-F screening to yield a compact, informative descriptor set. Across four held-out coating motifs (A–D), direct depth-wise hardness prediction with tree-based ensemble learners (XGBRegressor, CatBoost, and LGBM) achieves R^2≈ 0.72 with MAE≈ 1.6 . When the predicted profiles are subsequently processed using within-layer neighborhood averaging, the corresponding post-processed smoothed-profile agreement reaches R^2≈ 0.74 and MAE≈ 1.4 . This post-hoc operation attenuates short-range fluctuations while preserving systematic layer-scale trends; it does not modify the fitted regressors or the original experimental hardness values. When reformulated at the design-relevant resolution of layer means, ensemble models recover layer-level hardness with mean-by-layer R^2 up to ≈ 0.88 and MAE<0.81 on unseen motifs, supporting robust comparison of coating designs. Mechanistic interpretability is provided by SHAP, augmented with hierarchical clustering and graph-community analysis, which highlights coherent feature groups associated with depth transitions and phase-dependent responses. Scientific contribution statement: We introduce an integrated learning pipeline that couples continuous-depth nanoindentation representation with physics-aware descriptors to predict hardness in heterogeneous EBC systems while explicitly accounting for layer composition and stack position. In this framework, layer information is not inferred through a separate classification or segmentation task; instead, categorical layer descriptors are decoded into continuous phase-fraction features and used as model inputs. By demonstrating generalization across unseen coating motifs and by improving design-level (layer-mean) accuracy while retaining depth-wise resolution, the framework enables coating-scale mechanical profiling that is resilient to defects and local variability. Finally, we provide a structured interpretability workflow (SHAP + clustering + graph communities) that links predictive signals to physically meaningful descriptor families, supporting mechanistic insight and materials design decisions.
Identifying off-target interactions of approved drugs is important to anticipate side effects and uncover repurposing opportunities. Computational pipelines combining structural homology, structure prediction, and molecular dynamics (MD) simulations offer a promising strategy, but it remains unclear whether stable, control-like MD trajectories reliably indicate functional engagement. We examined this in a case study of two approved drugs. Using the Evolutionary Classification of Protein Domains (ECOD) framework to select candidate off-targets, we modeled each drug-protein complex as two independent AlphaFold3 models and simulated both by MD, for Seladelpar (a PPARδ agonist) and Zanamivir, an influenza neuraminidase inhibitor that also inhibits human Sialidase-2 (NEU2). Candidates were ranked by the similarity of global MD descriptors to the on-target control. For Seladelpar, the three top-ranked candidates (FXR, RARγ, ERRγ) were tested experimentally; the Zanamivir set was analyzed computationally only. None showed measurable activity in reporter or thermal shift assays, despite stable simulations and descriptor values comparable to the control. Including PPARα and PPARγ as weak-positive comparators, these descriptors did not rank genuine interactions closer to the control than inactive candidates. Residue-level comparison with experimental structures showed the predicted poses reproduced only part of the canonical contacts. Where experimental drug-bound structures existed, AlphaFold3 reproduced the pose for PPARα but not PPARγ, and its per-model confidence did not track pose accuracy. Within this case study, the specific global descriptors examined reflect complex stability rather than functional engagement, which does not mean MD-based approaches cannot make this distinction.
Molecular property prediction is a central task in drug discovery, yet acquiring labeled data remains costly and time-consuming. Self-supervised contrastive learning provides a promising route for learning from abundant unlabeled molecular data. However, current contrastive methods still face two challenges: constructing chemically meaningful positive and negative pairs, and adapting shared molecular representations to task-specific property signals. To address these limitations, we propose a Property-Aware Contrastive Learning (PACL) framework for molecular property prediction with adaptive substructures. PACL introduces a cross-scale contrastive learning strategy that aligns atomic-level representations with adaptively partitioned substructure-level embeddings, avoiding reliance on data augmentation or 3D conformer generation. During fine-tuning, task-specific learnable prototypes and a property-aware embedding module recalibrate molecular representations to emphasize property-relevant features. Evaluated on nine MoleculeNet benchmarks under scaffold splitting, PACL achieves the best average ROC-AUC for classification and the lowest average RMSE for regression among the compared methods. Visualization and prototype preference analyses further indicate that PACL separates structurally similar molecules with different functional properties more clearly in task-aligned representation spaces. These results support adaptive substructure contrastive learning and task-specific semantic alignment as an effective strategy for multi-task molecular property prediction in data-limited settings.
The rising incidence of antifungal resistance has underscored the need for novel strategies that can expedite the discovery of structurally diverse and optimized antifungal candidates. Conventional approaches to modify azoles are limited by narrow chemical space and problems related to resistance with the fungal CYP51 (Eeg11). The objective of this study was to develop a TRIZ-guided computational framework for unbiased generation, prioritization and evaluation of novel azole-like antifungal candidates targeting fungal CYP51. A set of TRIZ inventive principles was translated into medicinal chemistry-guided structural modification rules, to build a Python-based molecular design workflow. The workflow produced a virtual library of seventy azole-like candidates that were computationally filtered and ranked without manual candidate selection. The top-ranked candidates were subjected to a combined in silico pipeline involving SwissDock-based molecular docking against a heme-containing holo-Candida auris CYP51 model, ligand efficiency analysis, SwissADME pharmacokinetic prediction, RDKit-based quantitative estimate of drug-likeness (QED), ProTox-3.0 toxicity profiling, molecular interaction analysis, target-space profiling, and ASKCOS retrosynthetic assessment. The automated TRIZ-guided approach could generate azole-like scaffolds with chemical relevance for ready downstream evaluation. The most promising compound identified from the screening of the candidates was AZ-TRIZ-02, which showed better predicted binding affinity to CYP51 than fluconazole. Ligand efficiency analysis showed greater binding contribution per heavy atom, implying that the improved interaction was not due to increased molecular size. AZ-TRIZ-02 also showed good drug-like attributes, the pharmacokinetic properties were acceptable, and the preliminary toxicity profile was free of the hepatotoxicity, mutagenicity, carcinogenicity, and cytotoxicity alerts. Interaction analysis confirmed stable placement within the CYP51 catalytic pocket. Retrosynthetic assessment supported theoretical synthetic feasibility. The work shows the potential of TRIZ-guided innovation, cheminformatics and AI-assisted molecular design combined as a reproducible framework for antifungal lead discovery. Future studies will be necessary to confirm the translational potential of the identified candidate, including experimental synthesis and antifungal testing, CYP51 inhibition assays and overall safety assessment.
Bruton’s Tyrosine Kinase (BTK) has been recently recognized as an important drug design target for treating B-cell malignancies. Unfortunately, drug resistance is making the treatment of B-cell malignancies challenging. In this study, we employed a dual machine learning drug design approach that involves two parallel branches: the QSAR-GFA modeling (Branch 1) which engages the development of an interpretable structure activity relationships, and the KNIME® based machine learning modeling (Branch 2) which aimed at achieving high predictive accuracy. Subsequently, six machine learning modules, encompassing; the Random Forest model, the XGBoost Model (XGBoost), the Naive Bayes model, the k-Nearest Neighbors model, and the Support Vector Machines model were applied. Later, experimental validation through cytotoxicity assays against Raji lymphoma and K562 leukemia cell lines revealed several compounds that are exhibiting notable biological activity. For instance, compounds 166 and 168 which are acridine based displayed IC50 values of 0.087 and 4.92 µM in Raji cells, respectively. Also, hits 166 and 168 were subjected to enzymatic evaluation by using a radiometric HotSpot™ kinase assay against BTK and yielded an IC50 values of 12.2 µM and 23.8 µM, respectively. Finally, a key outcome of this study is the identification of a chemically distinct and novel, acridine-based antiproliferative hit candidates with measurable BTK inhibitory activity.
Lysine crotonylation (Kcr) is a vital posttranslational modification that plays a significant role in diverse biological processes such as DNA replication, the cell cycle, spermatogenesis, and embryonic stem cell differentiation. Abnormal Kcr levels are associated with multiple diseases including cancer, neurological disorders, and metabolic diseases, making it crucial for understanding disease pathogenesis and progression. In this study, we introduce ESM2-Kcr, a novel computational model designed for predicting Kcr sites. This model integrates the protein language model ESM2 with advanced deep learning techniques including LSTM, multi-head attention mechanism, and CNN. We elaborate on the specific feature extraction contributions of each deep learning module in ESM2-Kcr: LSTM captures the long-distance sequential dependency information of amino acid residues in protein sequences, CNN extracts the local short-range structural and functional features centered on lysine residues, and the multi-head attention mechanism adaptively assigns weight coefficients to key residue positions to highlight Kcr-related critical sequence information and filter out redundant noise. ESM2-Kcr encodes the protein sequences using ESM2_t30_150M_UR50D and further extracts comprehensive features. Through extensive comparisons with other protein language models like ProteinBERT and ProtT5, as well as different models within the ESM2 family, ESM2-Kcr demonstrates superior generalization ability and performance. The ESM2-Kcr model not only enhances our understanding of protein regulation but also holds great potential in identifying disease biomarkers and facilitating drug development. Future research directions may involve extending this framework to other posttranslational modifications. Data and codes are available at https://github.com/liukai23157/ESM2-Kcr .
Metabolites are important indicators of physiological and pathological states, and their alterations are closely associated with disease prediction. Therefore, prediction of metabolite-disease associations is valuable for disease research and biomarker discovery. Here, we propose DFFRGM, a dual-view fusion framework that integrates residual gated graph convolution and Mamba-based multi-hop dependency modeling for metabolite-disease association prediction. The framework jointly learns from homogeneous similarity networks and heterogeneous association networks, and then fuses the two views through bidirectional cross-attention. On two datasets, DFFRGM achieved AUC values of 98.54 and 98.89
Molecular docking is one of the most established methods in computational drug discovery, due to its balance of speed and accuracy. However, the accuracy of docking results depends on a number of different parameters, and systematic reference data for comparisons to more advanced methods for binding affinity prediction are still scarce. This study assesses the impact of key parameters on the accuracy of binding free energy estimates from docking, using nine benchmark systems with 278 high-affinity ligands. Using the Molecular Operating Environment (MOE), we evaluated combinations of three receptor structures (two crystal structures, one AlphaFold2 model), two force fields, two scoring functions, two receptor flexibility settings, and two statistical evaluation schemes. The performance of the docking approaches is measured based on the squared Pearson’s correlation coefficient (R²), the root mean square error (RMSE) with respect to the experimental binding affinities, as well as the mean signed error (MSE) and Kendall’s tau for individual targets and the full dataset. The results show that the scoring function and the protein structure are the most important factors for binding affinity accuracy in rigid docking with the MOE software. Amber10:EHT and MMFF94x force fields had the same average Rmean2 value, but Amber10:EHT had a lower average RMSEmean. AlphaFold2 protein models yielded lower binding affinity accuracy and higher errors compared to experimental crystal structures, although induced fit docking improved results. Using the original benchmark, we also compared several docking programs. DOCK6 and MOE performed best, with mean R² values of about 0.49 and 0.40, respectively. The remaining docking programs did not outperform a molecular weight regression baseline. For a subset of four targets (CDK2, JNK1, P38, TYK2) evaluated in previous work, the performance of the optimized DOCK6 and MOE protocols produced correlation coefficients similar to those reported for certain MM/PBSA, FMO, and Boltz2 implementations evaluated on the same target subset. This raises questions about potential dataset biases, the structural preparation, or the implementation of those methods. Docking therefore should be considered as an important and computationally inexpensive reference baseline for binding affinity prediction.
In this work, the geometrical, energetic and electronic properties of tetrameric self-assemblies formed by the monomers involving various isocytosine tautomers were investigated in both the gas phase and chloroform using the SMD solvation model at the M06-2X/6–311++G(d,p) level of theory. The results demonstrate that the thermodynamic stability order of the monomers containing different isocytosine tautomers is dependent on the phase in which they are investigated. Moreover, the calculated energies indicate that heteroleptic and homoleptic tetrameric structures composed of isocytosine units in the keto–amine tautomeric form are energetically preferred over those containing the keto–imine and enol–amine tautomers. Furthermore, homoleptic tetrameric assemblies show greater energetic preference than their heteroleptic counterparts. Several electronic properties, including band gap, first ionization energy, electron affinity, chemical potential, electrophilicity index, hardness, and softness were evaluated. The reduced band gap values observed for the heteroleptic macrocycles suggest their potential for conductivity-related applications, warranting further investigation of their charge-transport properties. The AIM and NBO analyses provided detailed insights into the strength and nature of hydrogen bonding interactions within the tetramers. Overall, these findings highlight the crucial influence of tautomeric forms on the structure, energetics, and electronic features of isocytosine-based assemblies, suggesting useful guidelines for designing functional nucleic acid analogues and advanced supramolecular materials.
The assignment of the stereochemical configuration of enantiomers is crucial for structure elucidation in drug discovery and molecular design. Machine learning models trained on experimental data can predict chiral observable properties and assist in configuration assignment by comparison with experimental results. We implemented scalar triple product (STP) descriptors with atomic properties from RDKit libraries to create chiral-atom-centered variants (cSTP) that more effectively encode molecular chirality. Using four high-resolution liquid chromatography datasets with different chiral stationary phases (Chiralpak AD-H, CROWNPAK CR(+), CROWNPAK CR-I(+), and Lux cellulose-1), we trained Random Forest models to predict enantiomer elution order. The cSTP descriptors, particularly those calculated within a single-bond sphere around chiral centers, performed closely to conventional Morgan fingerprints—superior to the latter in generalizability and with specific datasets. The best-performing cSTP_1 descriptors achieved 77–100
WEE1 kinase, a critical regulator of the G2/M checkpoint, represents a validated therapeutic target in tumors harboring defects in DNA damage response (DDR) pathways. Although clinical inhibitors such as adavosertib have demonstrated therapeutic potential, challenges, including selectivity constraints, dose-limiting toxicities, and emerging resistance, highlight the need to expand the structural diversity of WEE1-targeting chemotypes. Here, we report a scalable, machine-learning-integrated virtual screening framework designed to explore ultra-large chemical space spanning an input search space of approximately 884 million compounds from ZINC20 and 199,854 purchasable compounds from the SPECS database. Molecular representations using ECFP4 fingerprints combined with UMAP-based dimensionality reduction and K-means clustering enabled diversity-guided prioritization across distinct regions of chemical space. Multi-stage structure-based computational evaluation, including pharmacophore modelling, molecular docking, MM/GBSA rescoring, and 200-ns molecular dynamics simulations, yielded 18 high-confidence candidates, from which three structurally novel scaffolds were selected for detailed analysis. One ZINC-derived and two SPECS-derived compounds demonstrated predicted stable binding modes involving key WEE1 active-site residues and favourable estimated developability profiles. Critically, in silico selectivity profiling against the off-target PLK1 revealed structurally grounded differential binding, providing a computational basis for selectivity. Preliminary in vitro evaluation of one SPECS-derived compound (AJ-292/13095349) demonstrated antiproliferative activity in triple-negative breast cancer (TNBC) models. Collectively, this study establishes an efficient computational hit identification framework for ultra-large screening and reports structurally distinct starting points for WEE1-targeted oncology drug discovery. AI-Guided screening of ultra-large chemical libraries identifies structurally distinct WEE1 inhibitor candidates evaluated by molecular simulations and preliminary in vitro assays
Nonylphenol (NP), a widespread environmental endocrine disruptor, adversely affects male reproductive health through oxidative stress and impaired steroidogenesis. Although natural products have shown protective effects against reproductive toxicity, the bioactive compounds responsible for these effects and their molecular mechanisms remain inadequately characterized. This study investigated the protective potential of Cucurbita maxima seed extract (CMSE) against NP-induced reproductive toxicity in adult male albino mice and identified key phytochemicals associated with its activity. Male mice were divided into six groups: Control, NP-treated, CMSE (HD), NP+CMSE (LD), NP+CMSE (HD), and NP + CC, and treated for 35 days. LC-MS analysis identified eleven bioactive compounds in CMSE, which were subsequently evaluated through molecular docking and molecular dynamics simulations against key steroidogenic enzymes. Among the identified constituents, Ganoderic acid D exhibited the highest binding affinity and the most stable interactions with steroidogenic targets, suggesting a potential role in regulating testosterone biosynthesis and reproductive function. In vivo analyses demonstrated that CMSE significantly attenuated NP-induced reproductive toxicity. Treatment improved gonadosomatic index, restored antioxidant defenses (GSH, SOD, and CAT), reduced lipid peroxidation (MDA), normalized testicular protein and cholesterol levels, enhanced 3β-HSD and 17β-HSD activities, and restored serum testosterone concentrations. Histopathological examination further confirmed substantial recovery of testicular architecture following CMSE administration. The integrated in silico and in vivo evidence provides novel mechanistic insights into the reproductive protective effects of Cucurbita maxima and highlights Ganoderic acid D as a promising natural therapeutic candidate against NP-induced male reproductive dysfunction.
Colorectal cancer (CRC) is one of the common malignant tumors of the gastrointestinal tract, encompassing both colon cancer and rectal cancer. This study explores the therapeutic potential of T. wilfordii in CRC through network pharmacology, molecular docking, and in vitro experiments. Active components of T. wilfordii were screened using oral bioavailability (OB ≥ 30
Myeloperoxidase (MPO) has shown promise as a therapeutic target due to its critical role in inflammatory mechanisms and cancer progression. Despite extensive research on MPO inhibitors, the lack of integrated computational–experimental workflows constrains the efficient identification and validation of biologically relevant candidates. Hence, in this study, a ligand-based pharmacophore model of MPO inhibitors was developed to identify crucial molecular features required for inhibition. A quantitative structure–activity relationship (QSAR) model was built using the Genetic Function Approximation (GFA) algorithm, and the model statistics were found to be statistically significant (R2 = 0.765, R2_adj = 0.733, R2_Pred = 0.672, LOF = 1.2). The generated and validated pharmacophore model was used virtually to screen 53,352 compounds, yielding five structurally distinct hits. These hits were subjected to in vitro cytotoxicity analysis using two cancer cell lines. Among all the tested compounds, BTB11556 showed the strongest cytotoxic activity, with an IC50 of 12.5 μM against the Kasumi-1 leukemia cell line, and was therefore considered the lead compound. A one-way ANOVA of the IC50 values for the active compounds in Kasumi-1 cells showed a statistically significant difference in cytotoxic potency (p < 0.001). This integrated computational and experimental approach highlights the importance of pharmacophore-guided virtual screening, combined with QSAR modeling, in the development of MPO inhibitors. The findings from molecular docking, 500-ns molecular dynamics simulations, MM-GBSA calculations, and alanine scanning analyses collectively corroborate a stable binding mode of BTB11556 within the MPO active site. These results support further investigation of BTB11556 as a candidate compound associated with MPO-targeted therapeutic strategies.
Transformer-based models such as ChemBERTa have demonstrated strong performance across a wide range of molecular prediction tasks by learning contextual representations from SMILES strings. Despite this success, the internal organization of the chemical and structural information learned by these models remains poorly understood. In this work, we investigate which chemical concepts, functional groups, and structural features are captured by ChemBERTa’s attention mechanisms across layers and heads. To address this question, we formulate the Attention Matrix Pattern Capture (AMPC) problem and introduce two complementary algorithms: AMPC-Chem-FG, which assesses whether attention patterns reflect predefined chemical concepts and functional groups, and AMPC-Struct, which evaluates the alignment between attention patterns and three-dimensional molecular structure represented by Coulomb matrices. Using a curated dataset derived from the ZINC database, our framework identifies specialized layer-head pairs that consistently emerge as statistical outliers and exhibit strong associations with chemically meaningful features. The results reveal specialized attention patterns related to ring structures, bond types, chirality, and functional groups. Furthermore, a layer-head pair achieves approximately 65
Voltage-gated sodium channel NaV1.7 is a key mediator of electrical excitability and signal transmission in peripheral nociceptors and has emerged as a highly attractive therapeutic target for the development of novel analgesic agents. However, the development of selective NaV1.7 inhibitors has been characterized by significant challenges, with repeated failures in clinical trials despite encouraging preclinical data. In this study, we developed and validated a series of ligand-based pharmacophore models (LBPMs) that can be useful for the discovery of novel NaV1.7 inhibitors with improved selectivity profiles. Using the VGSC database as our primary data source, we focused on sulfonamide-based inhibitors targeting the voltage-sensing domain IV (VSD-IV). Validation against active compounds and decoys demonstrated that most models achieved good discrimination performance with high areas under the curve (AUC) and strong enrichment factors. External validation using 22 inhibitors extracted from recent literature confirmed the models’ capability to identify novel NaV1.7 inhibitors. Virtual screening of 3 million commercially available compounds retrieved promising hits and known inhibitors, with molecular docking studies revealing binding modes consistent with established sulfonamide-based inhibitors. Experimental validation identified one compound with measurable selectivity for NaV1.7 over NaV1.5, providing preliminary support for the utility of the developed virtual screening workflow. In parallel, we developed NaV1.5 LBPMs to assess selectivity profiles and minimize potential cardiotoxic effects. Overall, our findings provide valuable computational tools and structural insights for the rational design of selective NaV1.7 inhibitors, offering important starting points for developing analgesics with reduced off-target effects.
The present study reports the isolation and characterization of secondary metabolites from the methanolic extract of Knema malayana leaves, including five polymethoxylated flavones 3′,4′,5,7-tetramethoxyflavone (1), 3′,4′,5′,5,7-pentamethoxyflavone (2), 4′-hydroxy-3′,5′,5,7-tetramethoxyflavone (3), 3-hydroxy-3′,4′,5′,5,7-pentamethoxyflavone (4), and 3,3′,5′,5,7-pentamethoxyflavone (5). Structural elucidation was performed using 1D and 2D nuclear magnetic resonance (NMR), IR, MS, and, in the case of compound (2), single-crystal X-ray diffraction. The isolated compounds were evaluated for their inhibitory activity against acetylcholinesterase (AChE) and 5-lipoxygenase (5-LOX). Among them, compound (2) exhibited the most notable dual inhibitory activity, with IC50 values of 10.2 μM (AChE) and 12.6 μM (5-LOX). Molecular docking supported these findings, with binding affinities of –10.5 kcal/mol (AChE) and –7.2 kcal/mol (5-LOX), revealing key hydrogen bonds and π–cation interactions within active site residues. Other flavonoids, particularly compounds (1) and (5), also displayed notable binding profiles. In silico absorption, distribution, metabolism, excretion, and toxicity (ADMET) evaluation using the Toxometris.ai platform indicated favorable drug-likeness and safety profiles, including non-mutagenic and non-cardiotoxic predictions, along with acceptable pharmacokinetic properties such as aqueous solubility, Caco-2 permeability, plasma protein binding, and microsomal stability. These findings were further supported by promising physicochemical and medicinal chemistry properties, including quantitative estimate of drug-likeness (QED) scores ranging from 0.65 to 0.75 and the absence of pan-assay interference compounds (PAINS) alerts, among other parameters. This study highlights the therapeutic relevance of K. malayana polymethoxylated flavones as promising dual-target AChE and LOX inhibitor with favorable safety and pharmacokinetic properties, supporting their potential development for neuroinflammatory disorders.
Odorant receptors (ORs) are essential components of the olfactory system in Rhynchophorus ferrugineus (red palm weevil), an invasive pest that relies on chemical cues to locate and infest host palms. However, the absence of experimentally resolved OR structures has restricted molecular‑level characterisation and constrained efforts to apply structure‑based approaches for understanding and disrupting its olfactory mechanisms. To address this gap, we implemented a high‑throughput structural bioinformatics workflow to identify, curate, and model 110 OR proteins from publicly available sequence databases and literature sources. Structural clustering, conserved‑motif analysis, and comparison with available insect OR structures enabled the selection of two representative receptors, RferOR18148 and RferOrco, which reflect key structural features of the broader receptor repertoire. Structural comparison and motif conservation analyses indicated shared structural patterns within a central cavity-like region, suggesting potential functional relevance in ligand interaction. Molecular dynamics simulations demonstrated that both receptors maintain stable structural conformations within a membrane environment, with consistent secondary structure retention and limited structural deviation over the simulation period. Additionally, literature evidence indicates that these receptors are expressed in chemosensory tissues, supporting their biological relevance. Overall, this study provides structural insights into the OR repertoire of R. ferrugineus and presents a systematic computational framework for the identification and prioritization of representative receptors for future ligand interaction studies and virtual screening efforts aimed at disrupting olfactory-driven host-seeking behaviour.
