Understanding the relationship between dietary compounds and epigenetic mechanisms is central to research in nutrigenomics and human health. Structured datasets that systematically integrate dietary, molecular, and epigenetic information are essential for reproducible and data-driven investigation in this domain. This dataset compiles curated associations between 653 unique dietary compounds, 585 food sources, and 80 epigenetic targets, extracted from 1,963 primary research articles identified through systematic searches in PubMed and Web of Science. Compound identification was performed via large language model-assisted extraction and subsequently standardized using PubChem, RDKit, and semantic clustering procedures. Binding affinity data were incorporated for human proteins with validated epigenetic function, prioritizing records with quantitative measures. Variables are harmonized and annotated using recognized chemical and biological ontologies. This resource supports research on nutriepigenomics, compound–mechanism associations, and food-derived bioactives, enabling hypothesis generation, validation studies, and data mining in nutritional, biomedical, and computational research contexts.
With Plasmodium falciparum devising means of evading antimalarial agents and in the absence of a vaccine that can be used across all age groups, malaria continues to ravage large parts of the tropical and sub-tropical regions of the world. Hence improved drugs with novel scaffolds are constantly needed to replenish the antimalarial chemotherapeutic arsenal. Medicinal plants have afforded some of the best antimalarials of which members of the alkaloid class are the ones that have made the most significant contributions. Protoberberine alkaloids are berberine-like alkaloids present in a number of medicinal plants traditionally used for the management of malaria. These alkaloids have been confirmed to have in vitro antiplasmodial activity. However, their development as antimalarials has been marred by their poor bioavailability. Among other features expected of drugs, understanding their mechanism of antiplasmodial action can spur research into designing of orally active and pharmacokinetically robust analogues. To investigate the anti-plasmodial activity of protoberberine alkaloids, Berberine and Palmatine were chosen for the present study. The mechanistic study of both alkaloids including intra-parasite localization was achieved by taking advantage of their inherent fluorescence property. Herein, we show that the protoberberine alkaloids Berberine and Palmatine, interact with vacuolar P-glycoprotein homologue-1 (Pf Pgh-1) transporter, enter the digestive vacuole and inhibit heme to hemozoin detoxification leading to death of the parasite.
Alzheimer’s disease (AD) is a multifactorial neurodegenerative disorder for which current approved therapies provide temporary symptomatic relief, and no disease-modifying treatments are approved. Given the complex pathology of AD and the growing interest in multitarget therapeutic strategies, we performed structure-based virtual screening to identify phytochemicals capable of modulating multiple AD-related targets. A virtual library of 2,847 phytochemical compounds was docked against ten key proteins implicated in AD identified by literature mining, including: amyloid beta (Aβ), acetylcholinesterase (AChE), apolipoprotein E4 (APOE4), β-secretase (BACE1), butyrylcholinesterase (BChE), brain derived neurotrophic factor (BDNF), cyclin dependent kinase 5 (CDK5), glycogen synthase kinase three beta (GSK3β) and presenilin 1/2 (PSEN1/2). Docking results showed potential strong binding of 96 compounds at multiple key targets, and molecular dynamics simulations of the representative compounds Lactucopicrin, Oleanolic acid and Rutecarpine in complex with AChE and GSK3β demonstrated stable protein-ligand complexes, comparable to the known AChE-ligand, Donepezil, and the GSK3β-ligand, QQA. Post-docking filters based on multitarget and heavy atoms count narrowed the selection to 36 compounds. These 36 compounds were classified into families of Alkaloids (Rutecarpine, Sanguinarine, Strychnine), Steroidal alkaloids (Tomatidine, Cyclopamine, Jervine, Verticine, Verticinone), Steroidal sapogenins (Diosgenin, Hecogenin, Yamogenin, Tigogenin, Smilagenin, Sarsasapogenin, Neogitogenin, Ruscogenin), Triterpenoids/sterols (alpha-Spinasterin, Fucosterol, Stigmasterol, Beta-Sitosterol, gamma-Sitosterol, Isofucosterol, Germanicol, Lupeol, Beta-Amyrin, alpha-Amyrin, Taraxerol, Friedelin, Tetrahymanol, Tingenone, Parkeol, Methylenecycloartanol, Ursolic Acid, Oleanolic Acid, Obacunone), and Sesquiterpene (Lactucopicrin). Multiple compounds were supported by the literature, which shows that they can inhibit amyloid aggregation, tau phosphorylation, neuroinflammation, and cholinergic dysfunction, thus lending credibility to our claims. Collectively, our findings suggest that several of the identified multitarget phytochemicals are promising candidates for further experimental validation in AD therapy.
Günther K. H. Zupanc, Abigail L. Zodl, Naya M. Sabha
Drug discovery notoriously suffers from high attrition rates during clinical trials. A major contributing factor to this issue is the frequent failure to identify adverse neural functional effects of compounds in the CNS at earlier stages of drug development than currently possible. To address this unmet need, the (side-)effects of the widely used antitussive dextromethorphan were assessed employing a neuro-behavioral assay originally developed for evaluation of anesthetics. This high-content in vivo assay provides quadrimodal readouts, reflecting behavioral changes of a weakly electric fish model induced by the compound to be tested. Two of these readouts are proxies of the neural activity of the pacemaker nucleus, an endogenous oscillator that drives the periodic electric organ discharges. Administration of dextromethorphan induced a decrease in the frequency of these discharges, an increase in the generation of amplitude-frequency modulations of this electric behavior, and the production of an abnormal motor behavior, consisting of full-transverse-plane rotations of the fish. The rate of occurrence of each of these behaviors was dose dependent. The changes in discharge frequency and rotation of the fish’s body persisted for several hours and in some individuals even for at least 2 days. We hypothesize that the drug-induced changes in the two electric behaviors are mediated by inhibition of N -methyl-D-aspartate (NMDA) receptors and by activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)-receptors in the pacemaker nucleus. The results of this study serve as proof of concept for the applicability of the neuro-behavioral assay for the evaluation of (side-) effects of drug candidates on neural function in the CNS.
William J. Waldock, Ahmad Guni, Ara Darzi, Cesar de la Fuente-Nunez, Hutan Ashrafian
Peptides represent promising therapeutic agents due to their high specificity, biocompatibility, and capacity to modulate protein–protein interactions. However, the field faces critical challenges: inconsistent evaluation metrics, heterogeneous datasets, and poor reproducibility, which together undermine objective model comparison and benchmarking. To systematically review and quantitatively assess the predictive performance of machine learning (ML)-based peptide–target interaction (PTI) models, with emphasis on commonly reported metrics including area under the curve (AUC), concordance index (CI), and Precision. We conducted a systematic literature search across PubMed, arXiv, and Cochrane databases for studies published through 1 August 2025. Inclusion criteria required original ML-based peptide–target prediction models with quantitative performance metrics. Twenty-three studies met inclusion criteria. Fourteen reported AUC values (pooled estimate: 0.87, 95% Confidence Interval: 0.83–0.90), five reported Concordance Index values (0.90, 95% Confidence Interval: 0.89–0.91), and six reported precision (0.75, 95% Confidence Interval: 0.69–0.81). Top-performing models predominantly employed transformer or graph neural network architectures with structural input features. Critical limitations included inconsistent reporting practices, infrequent external validation, and limited data/code availability. The observation that the twenty-three studies present challenges to a pooled estimate is itself a consequence of the reporting practices we document, and is why we advance STRIDE. ML models demonstrate strong potential for PTI prediction, with leading approaches achieving robust classification and ranking performance. Nevertheless, progress is hindered by non-standardised evaluation metrics, limited transparency, and insufficient reproducibility. We introduce the STRIDE framework (encompassing Standardization, Transparency, Representativeness, Integration, Discovery, and Evidence) to establish rigorous evaluation standards, enhance methodological reproducibility, and support future evaluation of clinical applicability.
