Database Of Molecules Research Articles

In silico design, modelling and molecular mechanisms of Axl receptor tyrosine kinase inhibitors

A kinase domain from receptor tyrosine kinases (RTKs) regulate intracellular communications to control cellular metabolic activities. Some of the malignant cells have upregulated and overexpressed RTKs which are responsible for angiogenesis in many metastatic cancers. Axl RTK is present in most of the eukaryotic cells and all metastatic cancer cells have overexpressed Axl tyrosine kinase to trigger uncontrolled growth and angiogenesis in the malignant cells. The upregulated kinases can be inhibited in its active and inactive states in the presence of small organic molecule inhibitors. Kinase inhibitors have been discovered to arrest the signal transduction pathways in the malignant cells as a therapy and cure for cancer. In this work, small molecule databases were screened using the pharmacophore features of a macrocyclic inhibitor (7YS) taken as reference from the crystal structure of Axl kinase domain. Pharmacophore based virtual screening of small molecule libraries (CHEMBL32, ChemDiv, Chemspace, Mcule, MolProt, PubChem and Zinc), followed by molecular docking, molecular dynamics simulations, binding energies from MM-PBSA calculations and trajectory analysis as principal component analysis were studied. The molecular basis for the binding of macrocyclic inhibitor, ATP and seven screened hit molecules bound at Axl kinase domain in two different modes at catalytic and regulatory sites was analyzed.

A Deep Dive into PDE5 Inhibition: Innovative Discoveries via Virtual Screening

Background: PDE5 inhibitors have had a surge in popularity over the last decade owing to their efficacy in the treatment of erectile dysfunction, coronary vasculopathy, and pulmonary arterial hypertension. These inhibitors exhibit competitive binding with phosphodiesterase type 5 and inhibit the hydrolysis of cyclic guanosine monophosphate, hence elevating the levels of cGMP in smooth muscle cells and prolonging the duration of an erection. However, due to production costs and side effects, further research is needed to discover new PDE5 inhibitors. Objective: The study aimed to identify potent PDE5 inhibitors by employing the extensive application of computer-aided drug design. Methods: Three different databases, named Million Molecules Database, Natural Product Database, and NCI Database, have been screened, which has been followed by filtering based on various druglikeness rules, docking, ADME, toxicity, consensus molecular docking, and 100 ns molecular dynamics simulation. Results: Three compounds (ZINC05351336, ZINC12030898, and ZINC17949426) have exhibited stable-binding characteristics at the active site of PDE5, demonstrating a robust binding affinity. These molecules have been found to possess drug-like capabilities, effective ADME features, low toxicity, and high stability. Conclusion: The study has delved into the realm of PDE5 inhibitors, which have been found to be effective in treating erectile dysfunction, but high production costs and side effects necessitate new ones. Through computer-aided drug design and screening, three compounds have been identified with promising binding characteristics, drug-appropriate properties, effective ADME profiles, minimal toxicity, and stability, making them potential candidates for future PDE5 inhibitors

Synthesis and characterization of machine learning designed TADF molecules

In this study, we present a novel approach to the development of thermally activated delayed fluorescence (TADF) molecules with potentials for organic light-emitting diode (OLED) applications, leveraging machine learning (ML) algorithms to guide the materials design process. Recognizing the imperative for high-efficiency, low-cost emissive materials, we integrated ML driven models with experimental characterization to expedite the discovery of TADF compounds. Initially, a database of ML-designed TADF molecules was employed, with samples being approved to possess optimized photophysical properties. The proposed molecules were synthesized using palladium-catalyzed coupling reactions. Subsequent characterization of these molecules utilized a suite of analytical methods, including nuclear magnetic resonance (NMR) spectroscopy, photoluminescence (PL) spectroscopy, and transient PL decay etc., to confirm their structural integrity and evaluate their performance metrics. The photophysical analysis revealed notable emission efficiencies and significant delayed fluorescence characteristics in solution phases, underscoring the potential of ML-designed TADF molecules. Theoretical validations, through quantum chemical calculations, corroborated the experimental findings, demonstrating the predictive power of our ML models. This interdisciplinary approach not only accelerates the pace of TADF molecule development but also provides a scalable framework for future material innovation especially in the OLED research field.

Performance of uncertainty-based active learning for efficient approximation of black-box functions in materials science.

Obtaining a fine approximation of a black-box function is important for understanding and evaluating innovative materials. Active learning aims to improve the approximation of black-box functions with fewer training data. In this study, we investigate whether active learning based on uncertainty sampling enables the efficient approximation of black-box functions in regression tasks using various material databases. In cases where the inputs are provided uniformly and defined in a relatively low-dimensional space, the liquidus surfaces of the ternary systems are the focus. The results show that uncertainty-based active learning can produce a better black-box function with higher prediction accuracy than that by random sampling. Furthermore, in cases in which the inputs are distributed discretely and unbalanced in a high-dimensional feature space, datasets extracted from materials databases for inorganic materials, small molecules, and polymers are addressed, and uncertainty-based active learning is occasionally inefficient. Based on the dependency on the material descriptors, active learning tends to produce a better black-box functions than random sampling when the dimensions of the descriptor are small. The results indicate that active learning is occasionally inefficient in obtaining a better black-box function in materials science.

Peptaloid: A Comprehensive Database for Exploring Peptide Alkaloid.

Peptaloid is the first dedicated database for peptide alkaloid molecules, a unique class of naturally derived compounds known for their structural diversity and significant biological activities. Despite their promising potential in drug discovery and therapeutic development, research on peptide alkaloids has been limited by the absence of a comprehensive and centralized resource. Fragmented data across various sources have posed a significant challenge, underscoring the need for a specialized database to facilitate more efficient research and application. Peptaloid addresses this critical gap by providing a database with over 161,000 peptide alkaloid entries, each detailed with structural, physicochemical, and pharmacological properties. By leveraging advanced computational tools and machine learning, Peptaloid generates ADMET profiles, aiding in identifying and optimizing therapeutic candidates. Designed for versatility, the database supports various applications beyond drug discovery, including ecology and material sciences. Peptaloid (as a specialized database for peptide alkaloids) will play a crucial role in innovation and collaboration across scientific disciplines. Peptaloid is accessible at https://peptaloid.niser.ac.in.

RadicalPy: A Tool for Spin Dynamics Simulations.

Radical pairs (electron-hole pairs, polaron pairs) are transient reaction intermediates that are found and exploited in all areas of science, from the hard realm of physics in the form of organic semiconductors, spintronics, quantum computing, and solar cells to the soft domain of chemistry and biology under the guise of chemical reactions in solution, biomimetic systems, and quantum biology. Quantitative analysis of radical pair phenomena has historically been successful by a few select groups. With this in mind, we present an intuitive open-source framework in the Python programming language that provides classical, semiclassical, and quantum simulation methodologies. A radical pair kinetic rate equation solver, Monte Carlo-based spin dephasing rate estimations, and molecule database functionalities are implemented. We introduce the kine-quantum method, a new approach that amalgamates classical rate equations, semiclassical, and quantum techniques. This method resolves the prohibitively large memory requirement issues of quantum approaches while achieving higher accuracy, and it also offers wavelength-resolved simulations, producing time- and wavelength-resolved magnetic field effect simulations. Model examples illustrate the versatility and ease of use of the software, including the new approach applied to the magnetosensitive absorption and fluorescence of flavin adenine dinucleotide photochemistry, spin-spin interaction estimation from molecular dynamics simulations on radical pairs inside reverse micelles, radical pair anisotropy inside proteins, and triplet exciton pairs in anthracene crystals. The intuitive interface also allows this software to be used as a teaching or learning aid for those interested in the field of spin chemistry. Furthermore, the software aims to be modular and extensible, with the aim to standardize how spin dynamics simulations are performed.

Designing potential lead compounds targeting aminoglycoside N (6′)-acetyltransferase in Serratia marcescens: A drug discovery strategy

Serratia marcescens is an opportunistic human pathogen that causes urinary tract infections, ocular lens infections, and respiratory tract infections. S. marcescens employs various defense mechanisms to evade antibiotics, one of which is mediated by aminoglycoside N-acetyltransferase (AAC). In this mechanism, the enzyme AAC facilitates the transfer and linkage of the acetyl moiety from the donor substrate acetyl-coenzyme A to specific positions on antibiotics. This modification alters the antibiotic's structure, leading to the inactivation of aminoglycoside antibiotics. In the current scenario, antibiotic resistance has become a global threat, and targeting the enzymes that mediate resistance is considered crucial to combat this issue. The study aimed to address the increasing global threat of antibiotic resistance in Serratia marcescens by targeting the aminoglycoside N-acetyltransferase (AAC (6′)) enzyme, which inactivates aminoglycoside antibiotics through acetylation. Due to the absence of experimental structure, we constructed a homology model of aminoglycoside N (6′)-acetyltransferase (AAC (6′)) of S. marcescens using the atomic structure of aminoglycoside N-acetyltransferase AAC (6′)-Ib (PDB ID: 1V0C) as a template. The stable architecture and integrity of the modelled AAC (6′) structure were analyzed through a 100 ns simulation. Structure-guided high-throughput screening of four small molecule databases (Binding, Life Chemicals, Zinc, and Toslab) resulted in the identification of potent inhibitors against AAC (6′). The hits obtained from screening were manually clustered, and the five hit molecules were shortlisted based on the docking score, which are observed in the range of −17.09 kcal/mol to −11.95 kcal/mol. These selected five molecules displayed acceptable pharmacological properties in ADME predictions. The binding free energy calculations, and molecular dynamics simulations of ligand bound AAC (6′) complexes represented higher affinity and stable binding. The selected molecules demonstrated stable binding with AAC (6′), indicating their strong potential to hamper the binding of aminoglycoside in the respective site. and thereby inhibit. This process mitigates enzyme mediated AAC (6′) activity on aminoglycosides and reverse the bactericidal function of aminoglycosides, and also this method could serve as a platform for the development of potential antimicrobials.

Tanshinone IIA delays liver aging by modulating oxidative stress.

Organ-specific aging is increasingly recognized for its research significance, with liver aging demonstrating particular relevance due to its central role in metabolism. We have pioneered the discovery that the expression of ESRRG in the liver positively correlates with age and have established its association with clinical characteristics, including hepatic edema. Our findings link liver aging to a shift in oxidative stress states, where ESRRG, a crucial nuclear receptor responsive to oxidative stress, may be modulated by various small molecules. Through virtual screening of a natural medicinal molecule database followed by further validation, we confirmed that the natural compound Tanshinone IIA mitigates oxidative stress-induced damage in the liver via the ESRRG/Cyp2e1 pathway, thus decelerating liver aging. Importantly, our study also explores the dynamic impact of Tanshinone IIA on ESRRG conformation, providing a profound understanding of its molecular interactions with ESRRG and laying a foundation for the rational design of small molecules based on natural compounds.

FlavorDB2: An updated database of flavor molecules.

Flavor is expressed through the interaction of molecules via gustatory and olfactory mechanisms. Knowing the pivotal role of flavor molecules in food and fragrances, the existence of a comprehensive repository becomes valuable. Such a repository encompasses the information of flavor molecules characterizing their flavor profile, chemical properties, regulatory status, consumption statistics, taste/aroma threshold values, reported uses in food categories, and synthesis. FlavorDB2 (https://cosylab.iiitd.edu.in/flavordb2/) is an updated database of flavor molecules with a user-friendly interface consisting of 25,595 flavor molecules. Among these, 2254 flavor molecules are associated with 936 natural ingredients from 34 categories. This repository simplifies the search for flavor molecules and their associated attributes. This platform offers multifaceted applications, including food pairing. FlavorDB2 serves as a standard repository of flavor compounds for researchers, chefs, and fragrance industry.

Reevaluating safety pharmacology respiratory studies within the ICH S7A core battery: A multi-company evaluation of preclinical utility and clinical translation

Optimization of ICH safety guideline studies for inclusion into regulatory submissions is critical for resource conservation, animal use reduction, and efficient drug development. The ICH S7A guidance for Safety Pharmacology (SP) studies adopted in 2001 identified the core battery of studies to evaluate the acute safety of putative pharmaceutical molecules prior to First in Human (FIH) trials. To assess the utility of respiratory studies in predicting clinical AE's, seven pharmaceutical companies pooled preclinical and clinical respiratory findings. A large database of novel molecules included all relevant data from standard S7A respiratory (n = 459) and FIH studies (n = 309). The data were analyzed with respect to the progression of these molecules, clinical adverse event reporting of these same molecules, and achieved exposures. These S7A respiratory assay findings had no impact on compound progression, and only 12 of 309 drug candidates were ‘positive’ preclinically and reported a respiratory-related AE in clinical trials (i.e. cough, dyspnea, etc.), an overall incidence rate of 3.9%. Contingency tables/statistics support a lack of concordance of these preclinical assays. Overall, our extensive analysis clearly indicated that the preclinical respiratory assay fails to provide any prognostic value for detecting clinically relevant respiratory adverse events.

Natural Products and Traditional Medicine: Investigating the Role of Natural Products and Traditional Medicine in Modern Pharmacology

Natural products and Traditional medicines have been in the use for effective management of diseases and care of human suffering for the last thousands of years and are depicted to contain a vast database of Bioactive molecules of immense therapeutic significance. The following paper’s purpose is to reflect on these practice in arts of pharmacology with regards to their history, origin, operation, and challenges in the modern world. These are plant compounds, animal, and minerals and how they have been used in former time and even in this 21st century. Modern trends and novel approaches suggest that new technologies in the genomics, proteomics and metabolomics will change the direction of the natural products research emphasizing molecular targeting as well as the individual approach towards patient treatment. It also talks about the problems of standardization and quality of the work done; therefore there is a great emphasis on the clinical trial they also go further to address rules that govern the application of such technologies. Secondly, the expansion in the usage of complementary medicine and integrative approach with natural substances in treating diseases, as well as attempts to educate doctors and other healthcare practitioners on the merits of traditional medicine. Hence, the focus of future prospects continues to be laid on advanced technologies and the role of global multilateralism in progress and regulation. The author has demonstrated a lot more about modern pharmacology within this review and is making a call for assimilation of the modern discovery with native knowledge for the benefit of mankind.

Geometric deep learning for molecular property predictions with chemical accuracy across chemical space

Chemical engineers heavily rely on precise knowledge of physicochemical properties to model chemical processes. Despite the growing popularity of deep learning, it is only rarely applied for property prediction due to data scarcity and limited accuracy for compounds in industrially-relevant areas of the chemical space. Herein, we present a geometric deep learning framework for predicting gas- and liquid-phase properties based on novel quantum chemical datasets comprising 124,000 molecules. Our findings reveal that the necessity for quantum-chemical information in deep learning models varies significantly depending on the modeled physicochemical property. Specifically, our top-performing geometric model meets the most stringent criteria for “chemically accurate” thermochemistry predictions. We also show that by carefully selecting the appropriate model featurization and evaluating prediction uncertainties, the reliability of the predictions can be strongly enhanced. These insights represent a crucial step towards establishing deep learning as the standard property prediction workflow in both industry and academia.Scientific contributionWe propose a flexible property prediction tool that can handle two-dimensional and three-dimensional molecular information. A thermochemistry prediction methodology that achieves high-level quantum chemistry accuracy for a broad application range is presented. Trained deep learning models and large novel molecular databases of real-world molecules are provided to offer a directly usable and fast property prediction solution to practitioners.

Combined structure-based virtual screening and machine learning approach for the identification of potential dual inhibitors of ACC and DGAT2

Acetyl-coenzyme A carboxylase (ACC) and diacylglycerol acyltransferase 2 (DGAT2) are recognized as potential therapeutic targets for nonalcoholic fatty liver disease (NAFLD). Inhibitors targeting ACC and DGAT2 have exhibited the capacity to reduce hepatic fat in individuals afflicted with NAFLD. However, there are no reports of dual inhibitors targeting ACC and DGAT2 for the treatment of NAFLD. Here, we aimed to identify potential dual inhibitors of ACC and DGAT2 using an integrated in silico approach. Machine learning-based virtual screening of commercial molecule databases yielded 395,729 hits, which were subsequently subjected to molecular docking aimed at both the ACC and DGAT2 binding sites. Based on the docking scores, nine compounds exhibited robust interactions with critical residues of both ACC and DGAT2, displaying favorable drug-like features. Molecular dynamics simulations (MDs) unveiled the substantial impact of these compounds on the conformational dynamics of the proteins. Furthermore, binding free energy assessments highlighted the notable binding affinities of specific compounds (V003–8107, G340–0503, Y200–1700, E999–1199, V003–6429, V025–4981, V006–1474, V025–0499, and V021–8916) to ACC and DGAT2. The compounds proposed in this study, identified using a multifaceted computational strategy, warrant experimental validation as potential dual inhibitors of ACC and DGAT2, with implications for the future development of novel drugs targeting NAFLD.

Transfer learning across different chemical domains: virtual screening of organic materials with deep learning models pretrained on small molecule and chemical reaction data

Machine learning is becoming a preferred method for the virtual screening of organic materials due to its cost-effectiveness over traditional computationally demanding techniques. However, the scarcity of labeled data for organic materials poses a significant challenge for training advanced machine learning models. This study showcases the potential of utilizing databases of drug-like small molecules and chemical reactions to pretrain the BERT model, enhancing its performance in the virtual screening of organic materials. By fine-tuning the BERT models with data from five virtual screening tasks, the version pretrained with the USPTO–SMILES dataset achieved R2 scores exceeding 0.94 for three tasks and over 0.81 for two others. This performance surpasses that of models pretrained on the small molecule or organic materials databases and outperforms three traditional machine learning models trained directly on virtual screening data. The success of the USPTO–SMILES pretrained BERT model can be attributed to the diverse array of organic building blocks in the USPTO database, offering a broader exploration of the chemical space. The study further suggests that accessing a reaction database with a wider range of reactions than the USPTO could further enhance model performance. Overall, this research validates the feasibility of applying transfer learning across different chemical domains for the efficient virtual screening of organic materials.Scientific contributionThis study verifies the feasibility of applying transfer learning to large language models in different chemical fields to help organic materials perform virtual screening. Through the comparison of transfer learning from different chemical fields to a variety of organic material molecules, the high precision virtual screening of organic materials is realized.

Associations between Suspected Adverse Drug Reactions of HMG-CoA Reductase Inhibitors and Polypharmacology Using a National Registry Approach

Aims: The aim of this study was to explore the suspected adverse drug reaction (ADR) data of five licensed statins in the UK: atorvastatin, fluvastatin, pravastatin, rosuvastatin, and simvastatin. A secondary aim was to determine if there are any associations between the polypharmacological properties of the statins and their associated muscle-related side effects. Methods: The chemical database of bioactive molecules with drug-like properties, European Molecular Biology Laboratory (ChEMBL), was used to obtain data on the pharmacological interactions of statins with human proteins. The Medicines and Healthcare Products Regulatory Agency’s (MHRA) Yellow Card scheme was used to obtain reports of suspected ADRs from 2018 to 2022. The OpenPrescribing database was used to obtain the prescribing rates for statistical interpretation. Results: The study found no significant difference between the statins association with suspected ADRs across all organ classes (X2, p > 0.05). Fluvastatin was found to have a higher incidence of ADRs/100,000 Rx across multiple system organ classes. Conclusions: No significant difference was found between the suspected ADR incidence of the statins across all system organ classes.

Computational drug repurposing effort for identifying novel hits for the treatment of diseases such as endometriosis, uterine fibroids, and prostate cancer.

This research aimed to identify potential drug compounds from the ZINC15 molecule database that could effectively treat GnRH1R-related diseases. The study utilized molecular docking and molecular dynamics methods to achieve this goal, which is crucial in drug repurposing research. The virtual screening process involved analyzing known drug compounds using molecular docking. Additionally, molecular dynamics simulations and MM-GBSA were employed to evaluate the stability of the complexes and determine the interactions between the compounds and protein structure. As a result, this study provides significant insights for treating diseases such as endometriosis, uterine fibroids, and prostate cancer related to GnRH1R. The study also involved designing new drugs and identifying necessary molecular scaffolds.

Abstract LB021: BAU-243: A novel BCL-2 inhibitor inducing autophagic cell death in glioblastoma, unveiled through multistep virtual screening, comprehensive in vitro and in vivo animal model evaluation

Abstract The excessive expression of antiapoptotic members in the B-cell leukemia/lymphoma-2 (BCL-2) family proteins emerges as a noteworthy oncogenic target in cancer cells. Employing a multistep virtual screening and filtering approach that integrates structure and ligand-based drug design, we aimed to identify novel and potent BCL-2 inhibitors from a sizable small molecule database, comprising over 210,000 compounds. Subsequently, seven selected molecules underwent rigorous biological activity tests in human cancer cell lines and Time-Resolved Fluorescence Resonance Energy Transfer (TR-FRET) assays. Four of these compounds exhibited notable inhibitory effects on cancer cell proliferation and led to an increase in apoptotic cell fractions. Among the identified compounds, BAU-243 displayed a substantial affinity for Bcl-2, demonstrating efficacy in reducing glioblastoma (GBM) cell proliferation, including cancer-initiating cells. Notably, in contrast to the selective Bcl-2 inhibitor ABT-199, BAU-243 induced autophagic cell death in GBM cells. In vivo investigations using orthotopic brain tumor models revealed a significant reduction in tumor growth with BAU-243 compared to both the vehicle group and animals treated with ABT-199. Computational modeling simulations, in concordance with in vitro findings, proposed that BAU-243 activates autophagic cell death by disrupting the Beclin 1:Bcl-2 complex. This positions BAU-243 as a promising small molecule for the treatment of GBM. Citation Format: Seyma Calis, Berna Dogan, Asuman Celebi, Ozlem Yapicier, Turker Kilic, Eda Tahir Turanli, Timucin Avsar, Serdar Durdagi. BAU-243: A novel BCL-2 inhibitor inducing autophagic cell death in glioblastoma, unveiled through multistep virtual screening, comprehensive in vitro and in vivo animal model evaluation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB021.

In-silico screening of phytomolecules against multiple targets for wound management.

Chronic wound healing, especially in burns, is a major medical challenge with limited treatments. This study employs computational tools to identify phytomolecules that target multiple pathways involved in wound healing. By utilizing shape analysis, molecular docking, and binding energy calculations, potential compounds are pinpointed,to address the growing problem of chronic wounds. Initially, a set of phytomolecules from the ZINC database of natural molecules was screened to find compounds with shapes similar to well-known wound healing phytomolecules like curcumin, chromogenic acid, gallic acid, and quercetin. The most promising phytomolecules identified through shape similarity were further studied through molecular docking studies on several key targets involved in wound healing, including TNF-α, FGF, and TGF-β. Among the tested phytomolecules, a ligand known as Fluorophenyl(5-(5-chloro-1-(2-fluorophenyl)-2-oxopentyl)-4,5,6,7-tetrahydrothieno[3,2c]pyridine-2-yl acetate) exhibited a strong affinity with favourable binding interactions for TNF-α ( - 7.1kcal/mole), FGF (-6.9kcal/mole), and TGF-β (-5.1kcal/mole). Another compound, 2,4 methoxybenzylidene-(-3)-oxo-2,3-dihydro-1-benzofuran-6-yl-4-methoxybenzoate, demonstrated a strong affinity with low binding energy for TNF-α ( - 6.8kcal/mole) and FGF ( - 7.0kcal/mole) targets. Isosakuranetin and Ermanin displayed moderate affinity for both TNF-α and FGF, with the highest affinity observed for the TGF-β target. These findings suggest that these identified phytomolecules hold promise as potential lead compounds for further structural modifications, with the goal of designing new molecules that can target multiple pathways involved in the wound healing process.

Identifying novel aryl hydrocarbon receptor (AhR) modulators from clinically approved drugs: In silico screening and In vitro validation

The aryl hydrocarbon receptor (AhR) functions as a vital ligand-activated transcription factor, governing both physiological and pathophysiological processes. Notably, it responds to xenobiotics, leading to a diverse array of outcomes. In the context of drug repurposing, we present here a combined approach of utilizing structure-based virtual screening and molecular dynamics simulations. This approach aims to identify potential AhR modulators from Drugbank repository of clinically approved drugs. By focusing on the AhR PAS-B binding pocket, our screening protocol included binding affinities calculations, complex stability, and interactions within the binding site as a filtering method. Comprehensive evaluations of all DrugBank small molecule database revealed ten promising hits. This included flibanserin, butoconazole, luliconazole, naftifine, triclabendazole, rosiglitazone, empagliflozin, benperidol, nebivolol, and zucapsaicin. Each exhibiting diverse binding behaviors and remarkably very low binding free energy. Experimental studies further illuminated their modulation of AhR signaling, and showing that they are consistently reducing AhR activity, except for luliconazole, which intriguingly enhances the AhR activity. This work demonstrates the possibility of using computational modelling as a quick screening tool to predict new AhR modulators from extensive drug libraries. Importantly, these findings hold immense therapeutic potential for addressing AhR-associated disorders. Consequently, it offers compelling prospects for innovative interventions through drug repurposing.

Transfer learning-enabled classification of drugs for Alzheimers Disease

Alzheimers disease (AD) is a neurodegenerative disease with a worldwide prevalence of 760.5 per 100,000 inhabitants. Environment factors and genetic predispositions can influence an individuals risk of developing AD. Recent research provided insights into the genetic mechanisms of AD, but a comprehensive database of drug-like molecules that can exacerbate or reduce AD risk remains unavailable. Here, we use machine learning to create a similarity map that reveals new putative drugs structurally similar to existing compounds known to be targeting pathways relevant in AD. We trained an autoencoder on a large drug database of over 14,000 drugs, with features derived from several modalities including molecular fingerprints. We then computed similarity scores based on these reduced dimensions. We show that our model is able to identify new compounds structurally similar to existing drugs linked to AD. We conclude that our model holds the potential to elucidate new compounds based on structural similarity and can be used to identify new drugs that affect critical pathways in AD.