Search Results (17)

Background/Objectives: Colombia harbors exceptional plant diversity, comprising over 31,000 formally identified species, of which approximately 6000 are classified as useful plants. Among these, 2567 species possess documented food and medicinal applications, with several traditionally utilized for managing febrile illnesses. Despite the global burden of dengue virus infection affecting millions annually, no specific antiviral therapy has been established. This study aimed to identify potential anti-dengue compounds from Colombian medicinal flora through machine learning-based quantitative structure–activity relationship (QSAR) modeling. Methods: An optimized XGBoost algorithm was developed through Bayesian hyperparameter optimization (Optuna, 50 trials) and trained on 2034 ChEMBL-derived activity records with experimentally validated anti-dengue activity (IC₅₀/EC₅₀). The model incorporated 887 molecular features comprising 43 physicochemical descriptors and 844 ECFP4 fingerprint bits selected via variance-based filtering. IC₅₀ and EC₅₀ endpoints were modeled independently based on their pharmacological distinction and negligible correlation (r = −0.04, p = 0.77). Through a systematic literature review, 2567 Colombian plant species from the Humboldt Institute’s official checklist were evaluated (2501 after removing duplicates and infraspecific taxa), identifying 358 with documented antiviral properties. Phytochemical analysis of 184 characterized species yielded 3267 unique compounds for virtual screening. A dual-endpoint classification strategy categorized compounds into nine activity classes based on combined potency thresholds (Low: pActivity ≤ 5.0, Medium: 5.0 < pActivity ≤ 6.0, High: pActivity > 6.0). Results: The optimized model achieved robust performance (Matthews correlation coefficient: 0.583; ROC-AUC: 0.896), validated through hold-out testing (MCC: 0.576) and Y-randomization (p < 0.01). Virtual screening identified 276 compounds (8.4%) with high predicted potency for both endpoints (“High-High”). Structural novelty analysis revealed that all 276 compounds exhibited Tanimoto similarity < 0.5 to the training set (median: 0.214), representing 145 unique Murcko scaffolds of which 144 (99.3%) were absent from the training data. Application of drug-likeness filtering (QED ≥ 0.5) and applicability domain assessment identified 15 priority candidates. In silico ADMET profiling revealed favorable pharmaceutical properties, with Incartine (pIC₅₀: 6.84, pEC₅₀: 6.13, QED: 0.83), Bilobalide (pIC₅₀: 6.78, pEC₅₀: 6.07, QED: 0.56), and Indican (pIC₅₀: 6.73, pEC₅₀: 6.11, QED: 0.51) exhibiting the highest predicted potencies. Conclusions: This systematic computational screening of Colombian medicinal flora demonstrates the untapped potential of regional biodiversity for anti-dengue drug discovery. The identified candidates, representing structurally novel chemotypes, are prioritized for experimental validation. Full article

Bioactive peptides have significant potential for applications in pharmaceuticals, the food industry, and cosmetics due to their wide spectrum of biological activities. However, their pronounced structural and functional heterogeneity complicates the classification and prediction of biological activity. This study uses data from the PepLab platform, comprising 2748 experimentally confirmed bioactive peptides distributed across 15 functional classes, including ACE inhibitors, antimicrobial, anticancer, antioxidant, toxins, and others. For each peptide, the amino acid sequence and key physicochemical descriptors are provided, calculated via the integrated DMPep module, such as GRAVY index, aliphatic index, isoelectric point, molecular weight, Boman index, and sequence length. The dataset exhibits class imbalance, with class sizes ranging from 14 to 524 peptides. An innovative methodology is proposed, combining descriptive statistical analysis, structural modeling via DEMATEL, and structural equation modeling with neural networks (SEM-NN), where SEM-NN is used to capture complex nonlinear causal relationships between descriptors and functional classes. The results of these dependencies are integrated into a multi-class machine learning model to improve interpretability and predictive performance. Targeted data augmentation was applied to mitigate class imbalance. The developed classifier achieved predictive accuracy of up to 66%, a relatively high value given the complexity of the problem and the limited dataset size. These results confirm that integrating structured dependency modeling with artificial intelligence is an effective approach for functional peptide classification and supports the rational design of novel bioactive molecules. Full article

Biomimetic chromatography is a powerful tool used in the pharmaceutical industry to characterize the physicochemical properties of molecules during early drug discovery. Some studies have indicated that biomimetic chromatography may also be useful for the evaluation of toxicologically relevant molecules. In this study, we evaluated the usefulness of the biomimetic chromatography approach for determining the lipophilicity, affinity to phospholipids, and bind to plasma proteins of selected organophosphate pesticides. Quantitative structure–retention relationship (QSRR) models were proposed to understand the structural features that influence the experimentally determined properties. ACD/labs, Chemicalize, and alvaDesc software were used to calculate theoretical descriptors. Multilinear regression was used as the regression type, and feature selection was supported by a genetic algorithm. The obtained QSRR models were validated internally and externally, and they demonstrated satisfactory performance with key statistical parameters ranged from 0.844 to 0.914 for R² and 0.696–0.898 for R²_ext, respectively, indicating good predictive ability. Full article

Reproductive toxicity poses significant risks to fertility and progeny health, making its identification in pharmaceutical compounds crucial. In this study, we conducted a comprehensive in silico investigation of reproductive toxic molecules, identifying three distinct categories represented by Dimethylhydantoin, Phenol, and Dicyclohexyl phthalate. Our analysis included physicochemical properties, target prediction, and KEGG and GO pathway analyses, revealing diverse and complex mechanisms of toxicity. Given the complexity of these mechanisms, traditional molecule-target research approaches proved insufficient. Support Vector Machines (SVMs) combined with molecular descriptors achieved an accuracy of 0.85 in the test dataset, while our custom deep learning model, integrating molecular SMILES and graphs, achieved an accuracy of 0.88 in the test dataset. These models effectively predicted reproductive toxicity, highlighting the potential of computational methods in pharmaceutical safety evaluation. Our study provides a robust framework for utilizing computational methods to enhance the safety evaluation of potential pharmaceutical compounds. Full article

Cyclodextrins are macrocyclic rings composed of glucose residues. Due to their remarkable structural properties, they can form host–guest inclusion complexes, which is why they are frequently used in the pharmaceutical, cosmetic, and food industries, as well as in environmental and analytical chemistry. This review presents the reports from 2011 to 2023 on the quantitative structure–activity/property relationship (QSAR/QSPR) approach, which is primarily employed to predict the thermodynamic stability of inclusion complexes. This article extensively discusses the significant developments related to the size of available experimental data, the available sets of descriptors, and the machine learning (ML) algorithms used, such as support vector machines, random forests, artificial neural networks, and gradient boosting. As QSAR/QPR analysis only requires molecular structures of guests and experimental values of stability constants, this approach may be particularly useful for predicting these values for complexes with randomly substituted cyclodextrins, as well as for estimating their dependence on pH. This work proposes solutions on how to effectively use this knowledge, which is especially important for researchers who will deal with this topic in the future. This review also presents other applications of ML in relation to CD complexes, including the prediction of physicochemical properties of CD complexes, the development of analytical methods based on complexation with CDs, and the optimisation of experimental conditions for the preparation of the complexes. Full article

Solubility is not only a crucial physicochemical property for laboratory practice but also provides valuable insight into the mechanism of saturated system organization, as a measure of the interplay between various intermolecular interactions. The importance of these data cannot be overstated, particularly when dealing with active pharmaceutical ingredients (APIs), such as dapsone. It is a commonly used anti-inflammatory and antimicrobial agent. However, its low solubility hampers its efficient applications. In this project, deep eutectic solvents (DESs) were used as solubilizing agents for dapsone as an alternative to traditional solvents. DESs were composed of choline chloride and one of six polyols. Additionally, water–DES mixtures were studied as a type of ternary solvents. The solubility of dapsone in these systems was determined spectrophotometrically. This study also analyzed the intermolecular interactions, not only in the studied eutectic systems, but also in a wide range of systems found in the literature, determined using the COSMO-RS framework. The intermolecular interactions were quantified as affinity values, which correspond to the Gibbs free energy of pair formation of dapsone molecules with constituents of regular solvents and choline chloride-based deep eutectic solvents. The patterns of solute–solute, solute–solvent, and solvent–solvent interactions that affect solubility were recognized using Orange data mining software (version 3.36.2). Finally, the computed affinity values were used to provide useful descriptors for machine learning purposes. The impact of intermolecular interactions on dapsone solubility in neat solvents, binary organic solvent mixtures, and deep eutectic solvents was analyzed and highlighted, underscoring the crucial role of dapsone self-association and providing valuable insights into complex solubility phenomena. Also the importance of solvent–solvent diversity was highlighted as a factor determining dapsone solubility. The Non-Linear Support Vector Regression (NuSVR) model, in conjunction with unique molecular descriptors, revealed exceptional predictive accuracy. Overall, this study underscores the potency of computed molecular characteristics and machine learning models in unraveling complex molecular interactions, thereby advancing our understanding of solubility phenomena within the scientific community. Full article

Drug discovery aims to keep fueling new medicines to cure and palliate many ailments and some untreatable diseases that still afflict humanity. The ADME/Tox (absorption, distribution, metabolism, excretion/toxicity) properties of candidate drug molecules are key factors that determine the safety, uptake, elimination, metabolic behavior and effectiveness of drug research and development. The predictive technique of ADME/Tox drastically reduces the fraction of pharmaceutics-related failure in the early stages of drug development. Driven by the expectation of accelerated timelines, reduced costs and the potential to reveal hidden insights from vast datasets, artificial intelligence techniques such as Graphormer are showing increasing promise and usefulness to perform custom models for molecule modeling tasks. However, Graphormer and other transformer-based models do not consider the molecular fingerprint, as well as the physicochemicals that have been proved effective in traditional computational drug research. Here, we propose an enhanced model based on Graphormer which uses a tree model that fully integrates some known information and achieves better prediction and interpretability. More importantly, the model achieves new state-of-the-art results on ADME/Tox properties prediction benchmarks, surpassing several challenging models. Experimental results demonstrate an average SMAPE (Symmetric Mean Absolute Percentage Error) of 18.9 and a PCC (Pearson Correlation Coefficient) of 0.86 on ADME/Tox prediction test sets. These findings highlight the efficacy of our approach and its potential to enhance drug discovery processes. By leveraging the strengths of Graphormer and incorporating additional molecular descriptors, our model offers improved predictive capabilities, thus contributing to the advancement of ADME/Tox prediction in drug development. The integration of various information sources further enables better interpretability, aiding researchers in understanding the underlying factors influencing the predictions. Overall, our work demonstrates the potential of our enhanced model to expedite drug discovery, reduce costs, and enhance the success rate of our pharmaceutical development efforts. Full article

Blood–brain barrier (BBB) permeability is an essential phenomena when considering the treatment of neurological disorders as well as in the case of central nervous system (CNS) adverse effects caused by peripherally acting drugs. The presented work contains statistical analyses and the correlation assessment of the analyzed group of active pharmaceutical ingredients (APIs) with their BBB-permeability data collected from the literature (such as computational log BB; Kp,uu,brain, and CNS+/− groups). A number of regression models were constructed in order to observe the connections between the APIs’ physicochemical properties in combination with their retention data from the chromatographic experiments (TLC and HPLC) and the indices of bioavailability in the CNS. Conducted analyses confirm that descriptors significant in BBB permeability modeling are hydrogen bond acceptors and donors, physiological charge, or energy of the lowest unoccupied molecular orbital. These molecular descriptors were the basis, along with the chromatographic data from the TLC in log BB regression analyses. Normal-phase TLC data showed a significant contribution to the creation of the log BB regression model using the multiple linear regression method. The model using them showed a good predictive value at the level of R² = 0.87. Models for Kp,uu,brain resulted in lower statistics: R² = 0.56 for the group of 23 APIs with the participation of k IAM. Full article

Drug penetration through biological barriers is an important aspect of pharmacokinetics. Although the structure of the blood-brain and blood-milk barriers is different, a connection can be found in the literature between drugs entering the central nervous system (CNS) and breast milk. This study was created to reveal such a relationship with the use of statistical modelling. The basic physicochemical properties of 37 active pharmaceutical compounds (APIs) and their chromatographic retention data (TLC and HPLC) were incorporated into calculations as molecular descriptors (MDs). Chromatography was performed in a thin layer format (TLC), where the plates were impregnated with bovine serum albumin to mimic plasma protein binding. Two columns were used in high performance liquid chromatography (HPLC): one with immobilized human serum albumin (HSA), and the other containing an immobilized artificial membrane (IAM). Statistical methods including multiple linear regression (MLR), cluster analysis (CA) and random forest regression (RF) were performed with satisfactory results: the MLR model explains 83% of the independent variable variability related to CNS bioavailability; while the RF model explains up to 87%. In both cases, the parameter related to breast milk penetration was included in the created models. A significant share of reversed-phase TLC retention values was also noticed in the RF model. Full article

Genetic diversity in crop plants is the conditio sine qua non for sustainable agriculture and long-term food security. Our research carried out the morphological, agronomic, and physico-chemical characterization and resistance to late blight of 35 tomato landraces from seven countries. These landraces have been approved and appear in the Official Catalog of Varieties. The International Plant Genetic Resources Institute (IPGRI) descriptors have been used to describe the tomato’s morphological and agronomic characteristics. For the physico-chemical characteristics, the dry matter, the pH, and the carotenoid content (lycopene, lutein, and β-carotene)) were analyzed. Carotenoids were monitored by high-performance liquid chromatography (HPLC). The results showed that the morphological diversity of landraces was very high. Three landraces of remarkable commercial value have shown increased resistance to late blight caused by Phytophthora infestans, one of the most damaging diseases of tomato. Also, six landraces had a lycopene content exceeding 100 µg/g sample. The carotenoid content ranged between 0.769 (Marmande-FR 166) and 140.328 mg kg⁻¹ FW (Răscruci). The landrace with the highest β carotene content was PT 308 with 65.499 mg kg⁻¹ FW, while the lowest values were registered for Marmande-FR 166 with 0.105 mg kg⁻¹ FW. The present study provides essential information on the morphological and agronomic qualities of these tomato landraces and their lycopene and other carotenoid content. The results are discussed in light of the importance of tomato landraces in meeting the preferences of different producers and consumers, the choice of the most suitable landraces for specific pedoclimatic conditions, and the supply of carotenoid pigment sources for the pharmaceutical industry. Our research responds to humanity’s great global challenges: preserving agricultural biodiversity, protecting the environment by identifying pest-resistant varieties, and also protecting consumer health by finding important sources of antioxidants. Full article

The solubility of active pharmaceutical ingredients is a mandatory physicochemical characteristic in pharmaceutical practice. However, the number of potential solvents and their mixtures prevents direct measurements of all possible combinations for finding environmentally friendly, operational and cost-effective solubilizers. That is why support from theoretical screening seems to be valuable. Here, a collection of acetaminophen and phenacetin solubility data in neat and binary solvent mixtures was used for the development of a nonlinear deep machine learning model using new intuitive molecular descriptors derived from COSMO-RS computations. The literature dataset was augmented with results of new measurements in aqueous binary mixtures of 4-formylmorpholine, DMSO and DMF. The solubility values back-computed with the developed ensemble of neural networks are in perfect agreement with the experimental data, which enables the extensive screening of many combinations of solvents not studied experimentally within the applicability domain of the trained model. The final predictions were presented not only in the form of the set of optimal hyperparameters but also in a more intuitive way by the set of parameters of the Jouyban–Acree equation often used in the co-solvency domain. This new and effective approach is easily extendible to other systems, enabling the fast and reliable selection of candidates for new solvents and directing the experimental solubility screening of active pharmaceutical ingredients. Full article

The complexation of biogenic molecules with metals is the widespread strategy in screening for new pharmaceuticals with improved therapeutic and physicochemical properties. This paper demonstrates the possibility of using simple QSAR modeling based on topological descriptors for chelates study. The presence of a relationship between the structure (J) and lipophilic properties (logP) of zinc complexes with amino acids, where two molecules coordinate the central atom through carboxyl oxygen and amino group nitrogen, and thus form a double ring structure, was predicted. Using a cellular biosensor model for Gly, Ala, Met, Val, Phe and their complexes Zn(AA)₂, we experimentally confirmed the existence of a direct relationship between logP and biological activity (Ea). The results obtained using topological analysis, Spirotox method and microbiological testing allowed us to assume and prove that the chelate complex of zinc with methionine has the highest activity of inhibiting bacterial biofilms, while in aqueous solutions it does not reveal direct antibacterial effect. Full article

Machine learning (ML) approaches are receiving increasing attention from pharmaceutical companies and regulatory agencies, given their ability to mine knowledge from available data. In drug discovery, for example, they are employed in quantitative structure–property relationship (QSPR) models to predict biological properties from the chemical structure of a drug molecule. In this paper, following the Second Solubility Challenge (SC-2), a QSPR model based on artificial neural networks (ANNs) was built to predict the intrinsic solubility (logS₀) of the 100-compound low-variance tight set and the 32-compound high-variance loose set provided by SC-2 as test datasets. First, a training dataset of 270 drug-like molecules with logS₀ value experimentally determined was gathered from the literature. Then, a standard three-layer feed-forward neural network was defined by using 10 ChemGPS physico-chemical descriptors as input features. The developed ANN showed adequate predictive performances on both of the SC-2 test datasets. Benefits and limitations of ML approaches have been highlighted and discussed, starting from this case-study. The main findings confirmed that ML approaches are an attractive and promising tool to predict logS₀; however, many aspects, such as data quality, molecular descriptor computation and selection, and assessment of applicability domain, are crucial but often neglected, and should be carefully considered to improve predictions based on ML. Full article

In the postgenomic age, rapid growth in the number of sequence-known proteins has been accompanied by much slower growth in the number of structure-known proteins (as a result of experimental limitations), and a widening gap between the two is evident. Because protein function is linked to protein structure, successful prediction of protein structure is of significant importance in protein function identification. Foreknowledge of protein structural class can help improve protein structure prediction with significant medical and pharmaceutical implications. Thus, a fast, suitable, reliable, and reasonable computational method for protein structural class prediction has become pivotal in bioinformatics. Here, we review recent efforts in protein structural class prediction from protein sequence, with particular attention paid to new feature descriptors, which extract information from protein sequence, and the use of machine learning algorithms in both feature selection and the construction of new classification models. These new feature descriptors include amino acid composition, sequence order, physicochemical properties, multiprofile Bayes, and secondary structure-based features. Machine learning methods, such as artificial neural networks (ANNs), support vector machine (SVM), K-nearest neighbor (KNN), random forest, deep learning, and examples of their application are discussed in detail. We also present our view on possible future directions, challenges, and opportunities for the applications of machine learning algorithms for prediction of protein structural classes. Full article

Plants synthesize a large number of natural products, many of which are bioactive and have practical values as well as commercial potential. To explore this vast structural diversity, we present PSC-db, a unique plant metabolite database aimed to categorize the diverse phytochemical space by providing 3D-structural information along with physicochemical and pharmaceutical properties of the most relevant natural products. PSC-db may be utilized, for example, in qualitative estimation of biological activities (Quantitative Structure-Activity Relationship, QSAR) or massive docking campaigns to identify new bioactive compounds, as well as potential binding sites in target proteins. PSC-db has been implemented using the open-source PostgreSQL database platform where all compounds with their complementary and calculated information (classification, redundant names, unique IDs, physicochemical properties, etc.) were hierarchically organized. The source organism for each compound, as well as its biological activities against protein targets, cell lines and different organism were also included. PSC-db is freely available for public use and is hosted at the Universidad de Talca. Full article