Alzheimers Dement. 2024 Dec;20 Suppl 6:e089290. doi: 10.1002/alz.089290.

ABSTRACT

BACKGROUND: The prohibitive costs of drug development for Alzheimer's Disease (AD) emphasize the need for alternative in silico drug repositioning strategies. Graph learning algorithms, capable of learning intrinsic features from complex network structures, can leverage existing databases of biological interactions to improve predictions in drug efficacy. We developed a novel machine learning framework, the PreSiBOGNN, that integrates muti-modal information to predict cognitive improvement at the subject level for precision medicine in AD.

METHOD: The graph neural network framework integrates four layers of input data including transcriptome, proteome, drug, and subject to connect to bipartite graphs in the United Kingdom Biobank (UKBB). Medication usage, clinical, and GWAS data were downloaded for 48187 subjects with first and second cognitive exams from the UKBB. The protein and transcriptome layers were constructed using the String database and gene coexpression networks generated from single nuclei RNA data (Sahelijo et al. 2022). Layers were connected by binary bipartite graphs constructed using drug information from the UniProt and DrugCentral databases. Sequential Graph Attention Networks convoluted embedded features generated by each layer in a hierarchical order: 1. gene-gene, 2. gene-protein, 3. protein-protein, 4. protein-drug, 5. drug-subj. Feature embeddings were decoded using a multilayer perceptron to predict cognitive improvement between the first and second cognitive exams. Two models compared methods of data aggregation. The first model follows a strict hierarchy, aggregating inter-layer data in a single direction (gene > protein > drug > subj). The second model allows for fluid message passing between inter-layers. We used 60% of the UKBB subjects for training and 30% for validation. We assessed the model with the greatest training accuracy using the remaining 10% of subjects.

RESULT: We observed that the performance of the strict message-passing model attained validation and test accuracy of 57.7% and 51.2%, respectively. Performance of the fluid message passing model improved prediction accuracy with 61.3% and 58.1% in the test set.

CONCLUSION: Our investigation suggests the feasibility of the PreSiBOGNN framework to infer cognitive improvement of existing drugs by integrating medication, multi-omics, and clinical data. Future work will focus on model optimizations and the integration of additional modalities including compound-specific fingerprint data.

PMID:39782415 | DOI:10.1002/alz.089290

Alzheimers Dement. 2024 Dec;20 Suppl 6:e090350. doi: 10.1002/alz.090350.

ABSTRACT

BACKGROUND: Alzheimer's disease (AD) presents challenges with its complex neurodegenerative mechanisms, leading to a high failure rate in clinical trials. While drug repositioning offers a cost-effective solution, the lack of a subtype-driven strategy hinders success. Previously, we defined genetic subtypes and their prioritized genes for each genetic subtype (Sahelijo et al., 2022). This study evaluated unsupervised learning algorithms to characterize existing compounds targeting prioritized genes for the genetic subtypes.

METHOD: Compounds in at least Phase 2 clinical trials were gathered from PubChem. Active compounds against subtype-specific genes were selected, and their structural data were transformed into pharmacophore fingerprints using ChemmineR. Unsupervised algorithms (Agglomerative Clustering, Ensemble Clustering, Gaussian Mixture Models, Bayesian Gaussian Mixture Models) were optimized using evaluation metrics (Calinski-Harabasz, Davies-Bouldin, Silhouette) to cluster compounds within each subtype with the optimal cluster number and optimal algorithm. The finalized clusters of compounds were evaluated using significance values and mean within-cluster Jaccard similarity scores and were characterized with target genes and mechanisms of action.

RESULT: Four of the nine genetic subtypes generated compound clusters, including 3 clusters using Agglomerative Clustering for Ast-M2 with 11 targets and 180 compounds, 2 clusters using BGMM for Ast-M9 with 14 targets and 341 compounds, and 4 clusters using Ensemble Clustering for Oli-M45 with 11 targets and 66 compounds and for Oli-M50 with 18 genes and 431 compounds. We observed common structural signatures between Ast-M2, Ast-M9, and Oli-M50 clusters, while Oli-M45 clusters did not share any signatures with other clusters. The most significant cluster was found for the Oli-M45 subtype, with a cluster significance value of 8.49 and the highest mean compound similarity score of 0.96. The top-ranked cluster primarily contained Vinblastine formulations-microtubule and tubulin polymerization inhibitors-targeting TUBA1A and TUBA1B.

CONCLUSION: We demonstrated a novel drug repositioning framework for AD using unsupervised learning algorithms, enabling precision medicine and subtype-driven repositioning. This framework will be implemented in our future software tools.

PMID:39782372 | DOI:10.1002/alz.090350

J Inherit Metab Dis. 2025 Jan;48(1):e12835. doi: 10.1002/jimd.12835.

ABSTRACT

Inborn errors of metabolism (IEMs) are rare genetic conditions with significant morbidity and mortality. Technological advances have increased therapeutic options, making it challenging to remain up to date. A centralized therapy knowledgebase is needed for early diagnosis and targeted treatment. This study aimed to identify all treatable IEMs through a scoping literature review, followed by data extraction and analysis according to the Treatabolome principles. Knowledge of treatable IEMs, therapeutic categories, efficacy, and evidence was integrated into the Inborn Errors of Metabolism Knowledgebase (IEMbase), an online database encompassing all IEMs. The study identified 275 treatable IEMs, 18% of all currently known 1564 IEMs, according to the International Classification of Inherited Metabolic Disorders. Disorders of fatty acid and ketone body metabolism had the highest treatability (67%), followed by disorders of vitamin and cofactor metabolism (60%), and disorders of lipoprotein metabolism (42%). The most common treatment strategies were pharmacological therapy (34%), nutritional therapy (34%), and vitamin and trace element supplementation (12%). Treatment effects were most commonly observed in nervous system abnormalities (34%), metabolism/homeostasis abnormalities (33%), and growth (7%). Predominant evidence sources included case reports with evidence levels 4 (48%) and 5 (12%), and individual cohort studies with evidence level 2b (12%). Our study generated the Metabolic Treatabolome 2024. IEMs are the largest group of monogenic disorders amenable to disease-modifying therapy. With drug repurposing efforts and advancements in gene therapies, this number will expand. IEMbase now provides up-to-date, comprehensive information on clinical and biochemical symptoms and therapeutic options, empowering patients, families, healthcare professionals, and researchers in improving patient outcomes.

PMID:39777714 | DOI:10.1002/jimd.12835

Theranostics. 2025 Jan 1;15(3):1122-1134. doi: 10.7150/thno.101358. eCollection 2025.

ABSTRACT

Rationale: Acute liver failure (ALF) is characterized by rapid hepatic dysfunction, primarily caused by drug-induced hepatotoxicity. Due to the lack of satisfactory treatment options, ALF remains a fatal clinical disease, representing a grand challenge in global health. Methods: For the drug repositioning to ALF of mesalamine, which is clinically approved for the treatment of inflammatory bowel disease (IBD), we propose a supramolecular prodrug nanoassembly (SPNs). Mesalamine is modified with a functional peptide of the FRRG sequence. The resulting mesalamine prodrugs form nanoassemblies solely through intermolecular interactions, ensuring high drug loading capacity and reducing the potential toxicity associated with the carrier materials of conventional nanoparticle systems. Results: In acetaminophen (APAP)-induced ALF mouse models, the SPNs predominantly accumulate in injured target tissues owing to the nanoparticles' propensity to target the liver. Subsequently, cathepsin B overexpressed in hepatocytes by drug-induced inflammation triggers the release of mesalamine from the nanoassemblies via enzymatic cleavage, resulting in remarkable therapeutic efficacy. Meanwhile, nonspecific drug release in healthy cells is inhibited due to their relatively lower cathepsin B expression, which helps prevent the exacerbation of the ALF by minimizing adverse events related to drug exposure. Conclusions: This study provides valuable insights into designing rational nanomedicine for repurposing mesalamine in ALF treatment, potentially inspiring further research to discover effective and safe therapeutic options for patients.

PMID:39776792 | PMC:PMC11700871 | DOI:10.7150/thno.101358

Fundam Clin Pharmacol. 2025 Feb;39(1):e13046. doi: 10.1111/fcp.13046.

NO ABSTRACT

PMID:39775992 | DOI:10.1111/fcp.13046

Pharm Res. 2025 Jan 7. doi: 10.1007/s11095-024-03811-1. Online ahead of print.

ABSTRACT

Metabolic alterations are commonly associated with various cancers and are recognized as contributing factors to cancer progression, invasion, and metastasis. Drug repurposing, a strategy in drug discovery, utilizes existing knowledge to recommend established drugs for new indications based on clinical data or biological evidence. This approach is considered a less risky alternative to traditional drug development. Metformin, a biguanide, is a product of Galega officinalis (French lilac) primarily prescribed for managing type 2 diabetes, is recognized for its ability to reduce hepatic glucose production and enhance insulin sensitivity, particularly in peripheral tissues such as muscle. It also improves glucose uptake and utilization while decreasing intestinal glucose absorption. Statins, first isolated from the fungus Penicillium citrinum is another class of medication mainly used to lower cholesterol levels in individuals at risk for cardiovascular diseases, work by inhibiting the enzyme 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, which is essential for cholesterol biosynthesis in the liver. Metformin is frequently used in conjunction with statins to investigate their potential synergistic effects. Combination of metformin and simvastatin has gathered much attention in cancer research because of its potential advantages for cancer prevention and treatment. In this review, we analyze the effects of metformin and simvastatin, both individually and in combination, on key cancer hallmarks, and how this combination affects the expression of biomolecules and associated signaling pathways. We also summarize preclinical research, including clinical trials, on the efficacy, safety, and potential applications of repurposing metformin and simvastatin for cancer therapy.

PMID:39775614 | DOI:10.1007/s11095-024-03811-1

Elife. 2025 Jan 8;12:RP92491. doi: 10.7554/eLife.92491.

ABSTRACT

There are thousands of Mendelian diseases with more being discovered weekly and the majority have no approved treatments. To address this need, we require scalable approaches that are relatively inexpensive compared to traditional drug development. In the absence of a validated drug target, phenotypic screening in model organisms provides a route for identifying candidate treatments. Success requires a screenable phenotype. However, the right phenotype and assay may not be obvious for pleiotropic neuromuscular disorders. Here, we show that high-throughput imaging and quantitative phenotyping can be conducted systematically on a panel of C. elegans disease model strains. We used CRISPR genome-editing to create 25 worm models of human Mendelian diseases and phenotyped them using a single standardised assay. All but two strains were significantly different from wild-type controls in at least one feature. The observed phenotypes were diverse, but mutations of genes predicted to have related functions led to similar behavioural differences in worms. As a proof-of-concept, we performed a drug repurposing screen of an FDA-approved compound library, and identified two compounds that rescued the behavioural phenotype of a model of UNC80 deficiency. Our results show that a single assay to measure multiple phenotypes can be applied systematically to diverse Mendelian disease models. The relatively short time and low cost associated with creating and phenotyping multiple strains suggest that high-throughput worm tracking could provide a scalable approach to drug repurposing commensurate with the number of Mendelian diseases.

PMID:39773880 | DOI:10.7554/eLife.92491

BMC Bioinformatics. 2025 Jan 7;26(1):5. doi: 10.1186/s12859-024-06032-w.

ABSTRACT

The process of new drug development is complex, whereas drug-disease association (DDA) prediction aims to identify new therapeutic uses for existing medications. However, existing graph contrastive learning approaches typically rely on single-view contrastive learning, which struggle to fully capture drug-disease relationships. Subsequently, we introduce a novel multi-view contrastive learning framework, named CDPMF-DDA, which enhances the model's ability to capture drug-disease associations by incorporating diverse information representations from different views. First, we decompose the original drug-disease association matrix into drug and disease feature matrices, which are then used to reconstruct the drug-disease association network, as well as the drug-drug and disease-disease similarity networks. This process effectively reduces noise in the data, establishing a reliable foundation for the networks produced. Next, we generate multiple contrastive views from both the original and generated networks. These views effectively capture hidden feature associations, significantly enhancing the model's ability to represent complex relationships. Extensive cross-validation experiments on three standard datasets show that CDPMF-DDA achieves an average AUC of 0.9475 and an AUPR of 0.5009, outperforming existing models. Additionally, case studies on Alzheimer's disease and epilepsy further validate the model's effectiveness, demonstrating its high accuracy and robustness in drug-disease association prediction. Based on a multi-view contrastive learning framework, CDPMF-DDA is capable of integrating multi-source information and effectively capturing complex drug-disease associations, making it a powerful tool for drug repositioning and the discovery of new therapeutic strategies.

PMID:39773275 | DOI:10.1186/s12859-024-06032-w

J Transl Med. 2025 Jan 7;23(1):25. doi: 10.1186/s12967-024-06046-1.

ABSTRACT

BACKGROUND: Glioblastoma (GBM) is a rare brain cancer with an exceptionally high mortality rate, which illustrates the pressing demand for more effective therapeutic options. Despite considerable research efforts on GBM, its underlying biological mechanisms remain unclear. Furthermore, none of the United States Food and Drug Administration (FDA) approved drugs used for GBM deliver satisfactory survival improvement.

METHODS: This study presents a novel computational pipeline by utilizing gene expression data analysis for GBM for drug repurposing to address the challenges in rare disease drug development, particularly focusing on GBM. The GBM Gene Expression Profile (GGEP) was constructed with multi-omics data to identify drugs with reversal gene expression to GGEP from the Integrated Network-Based Cellular Signatures (iLINCS) database.

RESULTS: We prioritized the candidates via hierarchical clustering of their expression signatures and quantification of their reversal strength by calculating two self-defined indices based on the GGEP genes' log2 foldchange (LFC) that the drug candidates could induce. Among five prioritized candidates, in-vitro experiments validated Clofarabine and Ciclopirox as highly efficacious in selectively targeting GBM cancer cells.

CONCLUSIONS: The success of this study illustrated a promising avenue for accelerating drug development by uncovering underlying gene expression effect between drugs and diseases, which can be extended to other rare diseases and non-rare diseases.

PMID:39773231 | DOI:10.1186/s12967-024-06046-1

Postepy Biochem. 2024 Dec 2;70(4):438-446. doi: 10.18388/pb.2021_577.

ABSTRACT

Prostaglandyny są hormonami występującymi niemal we wszystkich ssaczych tkankach. Jako cząsteczki sygnałowe odgrywają one kluczową rolę w regulacji wielu procesów fizjologicznych, m. in. cyklu wzrostu włosa. W artykule opisano historię odkrycia prostaglandyn, w tym prace profesora Ryszarda Gryglewskiego - odkrywcy prostacykliny. Szczególną uwagę zwrócono na syntetyczny analog prostaglandyny F2α - latanoprost. Jest to lek wskazany w leczeniu jaskry, którego działaniem ubocznym jest nadmierny wzrost rzęs. Jako prolek, latanoprost ulega przekształceniu do aktywnego metabolitu - kwasu latanoprostowego. Dzięki ostatnim badaniom wiadomo, że kwas latanoprostowy ma szanse stać się skuteczną alternatywą dla minoksydylu i finasterydu - jedynych leków zarejestrowanych obecnie do leczenia łysienia androgenowego. Wprowadzenie na rynek leków przeciwko łysieniu zawierających pochodne prostaglandyn, w tym kwas latanoprostowy będzie procesem znacznie szybszym w porównaniu do tradycyjnej ścieżki rozwoju produktu, opartego o nowy związek chemiczny.

PMID:39772319 | DOI:10.18388/pb.2021_577

Pharmaceutics. 2024 Dec 17;16(12):1607. doi: 10.3390/pharmaceutics16121607.

ABSTRACT

Background/Objectives: Glucagon-like peptide-1 (GLP-1) receptor is currently one of the most explored targets exploited for the management of diabetes and obesity, with many aspects of its mechanisms behind cardiovascular protection yet to be fully elucidated. Research dedicated towards the development of oral GLP-1 therapy and non-peptide ligands with broader clinical applications is crucial towards unveiling the full therapeutic capacity of this potent class of medicines. Methods: This study describes the virtual screening of a natural product database consisting of 695,133 compounds for positive GLP-1 allosteric modulation. The database, obtained from the Coconut website, was filtered according to a set of physicochemical descriptors, then was shape screened against the crystal ligand conformation. This filtered database consisting of 26,325 compounds was used for virtual screening against the GLP-1 allosteric site. Results: The results identified ten best hits with the XP score ranging from -9.6 to -7.6 and MM-GBSA scores ranging from -50.8 to -32.4 and another 58 hits from docked pose filter and a second round of XP docking and MM-GBSA calculation followed by molecular dynamics. The analysis of results identified hits from various natural products (NPs) classes, to whom attributed antidiabetic and anti-obesity effects have been previously reported. The results also pointed to β-lactam antibiotics that may be evaluated in drug repurposing studies for off-target effects. The calculated ADMET properties for those hits revealed suitable profiles for further development in terms of bioavailability and toxicity. Conclusions: The current study identified several NPs as potential GLP-1 positive allosteric modulators and revealed common structural scaffolds including peptidomimetics, lactams, coumarins, and sulfonamides with peptidomimetics being the most prominent especially in indole and coumarin cores.

PMID:39771585 | DOI:10.3390/pharmaceutics16121607

Microorganisms. 2024 Dec 16;12(12):2602. doi: 10.3390/microorganisms12122602.

ABSTRACT

Toxoplasma gondii is a protozoan, and the etiologic agent of toxoplasmosis, a disease that causes high mortality in immunocompromised individuals and newborns. Despite the medical importance of toxoplasmosis, few drugs, which are associated with side effects and parasite resistance, are available for its treatment. Here, we show a screening of molecules present in COVID-Box to discover new hits with anti-T. gondii activity. COVID-Box contains 160 molecules with known or predicted activity against SARS-CoV-2. Our analysis selected 23 COVID-Box molecules that can inhibit the tachyzoite forms of the RH strain of T. gondii in vitro by more than 70% at 1 µM after seven days of treatment. The inhibitory curves showed that most of these molecules inhibited the proliferation of tachyzoites with IC50 values below 0.80 µM; Cycloheximide and (-)-anisomycin were the most active drugs, with IC50 values of 0.02 μM. Cell viability assays showed that the compounds are not toxic at active concentrations, and most are highly selective for parasites. Overall, all 23 compounds were selective, and for two of them (apilimod and midostaurin), this is the first report of activity against T. gondii. To better understand the effect of the drugs, we analyzed the effect of nine of them on the ultrastructure of T. gondii using transmission electron microscopy. After treatment with the selected drugs, the main changes observed in parasite morphology were the arrestment of cell division and organelle alterations.

PMID:39770804 | DOI:10.3390/microorganisms12122602

Pharmaceuticals (Basel). 2024 Nov 27;17(12):1601. doi: 10.3390/ph17121601.

ABSTRACT

Metformin is a commonly used drug for treating type 2 diabetes. Metformin is an inexpensive drug with low/no side effects and is well tolerated in human patients of different ages. Recent therapeutic strategies for human disease have considered the benefits of drug repurposing. This includes the use of the anti-diabetic drug metformin. Accordingly, the anti-inflammatory, anti-cancer, anti-viral, neuroprotective, and cardioprotective potentials of metformin have deemed it a suitable candidate for treating a plethora of human diseases. As results from preclinical studies using cellular and animal model systems appear promising, clinical trials with metformin in the context of non-diabetes-related illnesses have been started. Here, we aim to provide a comprehensive overview of the therapeutic potential of metformin in different animal models of human disease and its suggested relationship to epigenetics and ailments with epigenetic components.

PMID:39770443 | DOI:10.3390/ph17121601

Molecules. 2024 Dec 16;29(24):5932. doi: 10.3390/molecules29245932.

ABSTRACT

G-quadruplexes (G4s) are distinctive four-stranded nucleic acid structures formed by guanine-rich sequences, making them attractive targets for drug repurposing efforts. Modulating their stability and function holds promise for treating diseases like cancer. To identify potential drug candidates capable of interacting with these complex DNA formations, docking studies and molecular dynamics (MDs) simulations were conducted. Our analysis revealed kanamycin's ability to bind to various G4 structures, offering valuable insights into its potential as a modulator of G4 activity. Kanamycin exhibited favorable interactions with both parallel and hybrid G4 topologies in human structures, suggesting a broader mechanism of action for aminoglycosides. These findings may also shed light on aminoglycoside-associated toxicities, indicating that their effects might extend to binding non-ribosomal RNA structures. In summary, this research highlights kanamycin's potential as a promising tool for influencing G4 dynamics, paving the way for innovative therapeutic strategies targeting G4-related pathways.

PMID:39770021 | DOI:10.3390/molecules29245932

Int J Mol Sci. 2024 Dec 18;25(24):13573. doi: 10.3390/ijms252413573.

ABSTRACT

Finding an effective treatment for T-PLL patients remains a significant challenge. Alemtuzumab, currently the gold standard, is insufficient in managing the aggressiveness of the disease in the long term. Consequently, numerous efforts are underway to address this unmet clinical need. The rarity of the disease limits the ability to conduct robust clinical trials, making in silico, ex vivo, and in vivo drug screenings essential for designing new therapeutic strategies for T-PLL. We conducted a drug repurposing analysis based on T-PLL gene expression data and identified proteasome inhibitors (PIs) as a promising new class of compounds capable of reversing the T-PLL phenotype. Treatment of ex vivo T-PLL cells with Bortezomib and Carfilzomib, two PI compounds, supported this hypothesis by demonstrating increased apoptosis in leukemic cells. The current lack of a suitable in vitro model for the study of T-PLL prompted us to perform similar experiments in the SUP-T11 cell line, validating its potential by showing an increased apoptotic rate. Taken together, these findings open new avenues for investigating the molecular mechanisms underlying the efficacy of PI in T-PLL and expand the spectrum of potential therapeutic strategies for this highly aggressive disease.

PMID:39769335 | DOI:10.3390/ijms252413573

Biomolecules. 2024 Nov 27;14(12):1515. doi: 10.3390/biom14121515.

ABSTRACT

Neutrophil-mediated inflammation is a key feature of immune-mediated chronic skin disorders, but the mechanistic understanding of neutrophil involvement in these conditions remains incomplete. Dapsone, colchicine, and tetracyclines are established drugs within the dermatologist's therapeutic armamentarium that are credited with potent anti-neutrophilic effects. Anti-neutrophilic drugs have established themselves as versatile agents in the treatment of a wide range of dermatological conditions. Some of these agents are approved for the management of specific dermatologic conditions, but most of their current uses are off-label and only supported by isolated reports or case series. Their anti-inflammatory and immunomodulatory properties make them particularly valuable in managing auto-immune bullous diseases, neutrophilic dermatoses, eosinophilic dermatoses, interface dermatitis, and granulomatous diseases that are the focus of this review. By inhibiting inflammatory pathways, reducing cytokine production, and modulating immune responses, they contribute significantly to the treatment and management of these complex skin conditions. Their use continues to evolve as our understanding of these diseases deepens, and they remain a cornerstone of dermatological therapy.

PMID:39766222 | DOI:10.3390/biom14121515

Biomolecules. 2024 Nov 24;14(12):1497. doi: 10.3390/biom14121497.

ABSTRACT

Tuberculosis (TB) is one of the leading causes of death due to infectious disease. It is a treatable disease; however, conventional treatment requires a lengthy treatment regimen with severe side effects, resulting in poor compliance among TB patients. Intermittent drug use, the non-compliance of patients, and prescription errors, among other factors, have led to the emergence of multidrug-resistant TB, while the mismanagement of multidrug-resistant TB (MDR-TB) has eventually led to the development of extensively drug-resistant tuberculosis (XDR-TB). Thus, there is an urgent need for new drug development, but due to the enormous expenses and time required (up to 20 years) for new drug research and development, new therapeutic approaches to TB are required. Host-directed therapies (HDT) could be a most attractive strategy, as they target the host defense processes instead of the microbe and thereby may prevent the alarming rise of MDR- and XDR-TB. This paper reviews the progress in HDT for the treatment of TB using repurposed drugs which have been investigated in clinical trials (completed or ongoing) and plant-derived natural products that are in clinical or preclinical trial stages. Additionally, this review describes the existing challenges to the development and future research directions in the implementation of HDT.

PMID:39766204 | DOI:10.3390/biom14121497

Cancers (Basel). 2024 Dec 20;16(24):4253. doi: 10.3390/cancers16244253.

ABSTRACT

Despite significant progress in the field of clinical oncology in terms of diagnostic and treatment methods, the results of anticancer therapy are still not fully satisfactory, especially due to limited response and high toxicity. This has forced the need for further research to finding alternative ways to improve success rates in oncological treatment. A good solution to this problem in the context of rapidly obtaining an effective drug that works on multiple levels of cancer and is also safe is the global strategy of repurposing an existing drug. Research into other applications of an existing drug enables a precise assessment of its possible mechanisms of action and, consequently, the broadening of therapeutic indications. This strategy is also supported by the fact that most non-oncological drugs have pleiotropic effects, and most of the diseases for which they were originally intended are multifactorial, which in turn is a very desirable phenomenon due to the heterogeneous and multifaceted biology of cancer. In this review, we will mainly focus on the anticancer potential of H1 antihistamines, especially the new generation that were not originally intended for cancer therapy, to highlight the relevant signaling pathways and discuss the properties of these agents for their judicious use based on the characteristic features of cancer.

PMID:39766152 | DOI:10.3390/cancers16244253

Health Inf Sci Syst. 2025 Jan 5;13(1):14. doi: 10.1007/s13755-024-00326-2. eCollection 2025 Dec.

ABSTRACT

PURPOSE: Drug repositioning, a strategy that repurposes already-approved drugs for novel therapeutic applications, provides a faster and more cost-effective alternative to traditional drug discovery. Network-based models have been adopted by many computational methodologies, especially those that use graph neural networks to predict drug-disease associations. However, these techniques frequently overlook the quality of the input network, which is a critical factor for achieving accurate predictions.

METHODS: We present a novel network-based framework for drug repositioning, named DRAW+, which incorporates noise filtering and feature extraction using graph neural networks and attention mechanisms. The proposed model first constructs a heterogeneous network that integrates the drug-disease association network with the similarity networks of drugs and diseases, which are upgraded through reduced-rank singular value decomposition. Next, a subgraph surrounding the targeted drug-disease node pair is extracted, allowing the model to focus on local structures. Graph neural networks are then applied to extract structural representation, followed by attention walking to capture key features of the subgraph. Finally, a multi-layer perceptron classifies the subgraph as positive or negative, which indicates the presence of the link between the target node pair.

RESULTS: Experimental validation across three benchmark datasets showed that DRAW+ outperformed seven state-of-the-art methods, achieving the highest average AUROC and AUPRC, 0.963 and 0.564, respectively. Moreover, DRAW+ demonstrated its robustness by achieving the best performance across two additional datasets, further confirming its generalizability and effectiveness in diverse settings.

CONCLUSIONS: The proposed network-based computational approach, DRAW+, demonstrates exceptional accuracy and robustness, confirming its effectiveness in drug repositioning tasks.

PMID:39764174 | PMC:PMC11700073 | DOI:10.1007/s13755-024-00326-2

bioRxiv [Preprint]. 2024 Dec 16:2024.12.10.627863. doi: 10.1101/2024.12.10.627863.

ABSTRACT

Benchmarking is an important step in the improvement, assessment, and comparison of the performance of drug discovery platforms and technologies. We revised the existing benchmarking protocols in our Computational Analysis of Novel Drug Opportunities (CANDO) multiscale therapeutic discovery platform to improve utility and performance. We optimized multiple parameters used in drug candidate prediction and assessment with these updated benchmarking protocols. CANDO ranked 7.4% of known drugs in the top 10 compounds for their respective diseases/indications based on drug-indication associations/mappings obtained from the Comparative Toxicogenomics Database (CTD) using these optimized parameters. This increased to 12.1% when drug-indication mappings were obtained from the Therapeutic Targets Database. Performance on an indication was weakly correlated (Spearman correlation coefficient > 0.3) with indication size (number of drugs associated with an indication) and moderately correlated (correlation coefficient > 0.5) with compound chemical similarity. There was also moderate correlation between our new and original benchmarking protocols when assessing performance per indication using each protocol. Benchmarking results were also dependent on the source of the drug-indication mapping used: a higher proportion of indication-associated drugs were recalled in the top 100 compounds when using the Therapeutic Targets Database (TTD), which only includes FDA-approved drug-indication associations (in contrast to the CTD, which includes associations drawn from the literature). We also created compbench, a publicly available head-to-head benchmarking protocol that allows consistent assessment and comparison of different drug discovery platforms. Using this protocol, we compared two pipelines for drug repurposing within CANDO; our primary pipeline outperformed another similarity-based pipeline still in development that clusters signatures based on their associated Gene Ontology terms. Our study sets a precedent for the complete, comprehensive, and comparable benchmarking of drug discovery platforms, resulting in more accurate drug candidate predictions.

PMID:39764006 | PMC:PMC11702551 | DOI:10.1101/2024.12.10.627863