Compound Data Sets Research Articles

Measurement of ship-generated waves in German coastal waterways from 1998–2022

Vessel-generated waves and currents significantly impact coastal and estuarine waterways. In-situ measurements record all relevant physical phenomena that occur under a wide range of conditions and are therefore a valuable resource in the investigation of ship waves. Here we present a comprehensive compound dataset from in-situ ship wave measurement campaigns conducted over several decades in German coastal waterways. The dataset includes measured ship waves heights and currents as well as an attribution of each wave event to ship and nautical parameters responsible for its generation. The dataset is unique in its scope, combining a significant number of data points from multiple sites with different cross-section profiles. Some campaigns feature multiple gauges per cross-section, facilitating the analysis of wave transformation processes. Some locations feature repeat measurements which helps to understand wave load histories due to e.g. changes in shipping fleet structures and navigation practices. This dataset aims to provide a basis to enhance data-driven models to predict and manage the impacts of vessel-generated waves in different topographic settings and fleet structures.

Deep phenotypic profiling of neuroactive drugs in larval zebrafish

Behavioral larval zebrafish screens leverage a high-throughput small molecule discovery format to find neuroactive molecules relevant to mammalian physiology. We screen a library of 650 central nervous system active compounds in high replicate to train deep metric learning models on zebrafish behavioral profiles. The machine learning initially exploited subtle artifacts in the phenotypic screen, necessitating a complete experimental re-run with rigorous physical well-wise randomization. These large matched phenotypic screening datasets (initial and well-randomized) provide a unique opportunity to quantify and understand shortcut learning in a full-scale, real-world drug discovery dataset. The final deep metric learning model substantially outperforms correlation distance–the canonical way of computing distances between profiles–and generalizes to an orthogonal dataset of diverse drug-like compounds. We validate predictions by prospective in vitro radio-ligand binding assays against human protein targets, achieving a hit rate of 58% despite crossing species and chemical scaffold boundaries. These neuroactive compounds exhibit diverse chemical scaffolds, demonstrating that zebrafish phenotypic screens combined with metric learning achieve robust scaffold hopping capabilities.

Constraining light dependency in modeled emissions through comparison to observed biogenic volatile organic compound (BVOC) concentrations in a southeastern US forest

Abstract. Climate change will bring about changes in meteorological and ecological factors that are currently used in global-scale models to calculate biogenic emissions. By comparing long-term datasets of biogenic compounds to modeled emissions, this work seeks to improve understanding of these models and their driving factors. We compare speciated biogenic volatile organic compound (BVOC) measurements at the Virginia Forest Research Laboratory located in Fluvanna County, VA, USA, for the year 2020 with emissions estimated by the Model of Emissions of Gases and Aerosols from Nature version 3.2 (MEGANv3.2). The emissions were subjected to oxidation in a 0-D box model (F0AM v4.3) to generate time series of modeled concentrations. We find that default light-dependent fractions (LDFs) in the emissions model do not accurately represent observed temporal variability in regional observations. Some monoterpenes with a default light dependence are better represented using light-independent emissions throughout the year (LDFα-pinene=0, as opposed to 0.6), while others are best represented using a seasonally or temporally dependent light dependence. For example, limonene has the highest correlation between modeled and measured concentrations using an LDF =0 for January through April and roughly 0.74–0.97 in the summer months, in contrast to the default value of 0.4. The monoterpenes β-thujene, sabinene, and γ-terpinene similarly have an LDF that varies throughout the year, with light-dependent behavior in summer, while camphene and α-fenchene follow light-independent behavior throughout the year. Simulations of most compounds are consistently underpredicted in the winter months compared to observed concentrations. In contrast, day-to-day variability in the concentrations during summer months are relatively well captured using the coupled emissions–chemistry model constrained by regional concentrations of NOX and O3.

Identification of novel DNA gyrase inhibitor by combined pharmacophore modeling, QSAR analysis, molecular docking, molecular dynamics, ADMET and DFT approaches

DNA gyrase, an ATP-dependent enzyme, plays a critical role in DNA replication, transcription, and recombination in Mycobacterium tuberculosis (MTB). While fluoroquinolones are effective antibacterial agents targeting DNA gyrase, their clinical use is often limited due to side effects and the emergence of bacterial resistance. In this study, we developed a quantitative structure-activity relationship (QSAR) model to predict the anti-tubercular activity of Quinoline-Aminopiperidine derivatives targeting the DNA gyrase enzyme, using a dataset of 48 compounds obtained from the literature. The QSAR model was validated using both internal and external validation metrics. Model 4, the best predictive model, demonstrated a strong fit with an R² of 0.8393, an adjusted R² (R²adj) of 0.8010, and a lack of fit (LOF) parameter of 0.0626. The QSAR results revealed that DNA gyrase inhibition is significantly influenced by factors such as partition coefficient, molecular flexibility, hydrogen bonding potential, and the presence of fluorine atoms. Twelve quinoline-aminopiperidine derivatives were designed, and their anti-tubercular activity was predicted using QSAR model-4. These compounds were further assessed for pharmacokinetic properties, toxicity, and binding affinity to DNA gyrase. Pharmacophore modeling was also performed and validated using MOE software. The final pharmacophore model includes the features of two aromatic hydrophobic features, one hydrogen bond acceptor, and one hydrogen bond donor. The results indicated that designed compounds QA-3 and dataset compounds C-34 exhibit favorable drug-likeness properties. Molecular dynamics simulations confirmed the stable binding of compounds QA-3 and C-34 to the DNA gyrase protein, highlighting their potential as promising anti-tubercular agents. The developed QSAR Model-4 will facilitate the prediction of anti-tubercular activity in Quinoline-Aminopiperidine derivatives.

Application of Machine Learning for the Prediction of Absorption, Distribution, Metabolism and Excretion (ADME) Properties from Cichorium intybus Plant Phytomolecules

Drug discovery is the process by which new drug candidates are discovered and drug development takes place. To enhance the efficiency, accuracy, and speed of the drug discovery process, machine learning (ML) could play a transformative role. For this research study, antidiabetic natural compounds from C. intybus, which is commonly known as chicory, were selected, as they have promising antidiabetic properties that can complement conventional diabetes treatments. A bioactive natural compound dataset was retrieved on the chicory plant using Indian Medicinal Plants, Phytochemistry, and Therapeutics (IMPPAT) public source information. This collected dataset was analyzed for its absorption, distribution, metabolism, and excretion (ADME) properties using the SwissADME online tool. Principal component analysis (PCA) and correlation analysis were performed using trial-version XLSTAT software 2014.5.03 and Python. The obtained dataset from SwissADME was subjected to cleaning, after that, it was used to develop machine learning models, such as support vacuum (SVM) ML, random forest (RF), Naive Bayes (NB), and decision tree (DT). The Lipinski rule of violation was chosen as the target variable. To improve the vitality of the created ADME dataset, PCA, a biplot graph, and correlation analysis were carried out. A large dataset of naturally occurring antidiabetic compounds was used to predict the drug-likeness of ML models that were effectively deployed on heterogeneous ADME datasets. Among all these ML models, DT performed better than the rest of the models.

Word reuse and combination support efficient communication of emerging concepts

A key function of the lexicon is to express novel concepts as they emerge over time through a process known as lexicalization. The most common lexicalization strategies are the reuse and combination of existing words, but they have typically been studied separately in the areas of word meaning extension and word formation. Here, we offer an information-theoretic account of how both strategies are constrained by a fundamental tradeoff between competing communicative pressures: Word reuse tends to preserve the average length of word forms at the cost of less precision, while word combination tends to produce more informative words at the expense of greater word length. We test our proposal against a large dataset of reuse items and compounds that appeared in English, French, and Finnish over the past century. We find that these historically emerging items achieve higher levels of communicative efficiency than hypothetical ways of constructing the lexicon, and both literal reuse items and compounds tend to be more efficient than their nonliteral counterparts. These results suggest that reuse and combination are both consistent with a unified account of lexicalization grounded in the theory of efficient communication.

Structure‐Based Computational Approach in Search of the Potent Molecules Targeting Phosphoglycerate Dehydrogenase (PHGDH) Enzyme for Cancer Treatment

AbstractThis work aims to target phosphoglycerate dehydrogenase (PHGDH), a promising druggable target, that is overexpressed in various types of cancer. A structure‐based approach was employed to identify novel inhibitors against the enzyme. A common five‐feature pharmacophore model (RRHDA) was constructed using the active site co‐crystalized ligands. These chemical features were responsible for showing inhibition. The generated models were subsequently subjected to the validation method using a test set, receiver‐operator characteristic analysis, enrichment factor, and Güner–Henry studies. The validated models were subjected to the screening of a dataset of natural compounds. The screened unique natural compounds (1795) were further selected for the interaction analysis and study of ligand binding affinity considering the effect of the hydrogen bonding and desolvation and hydrophobic interactions contribution to the binding. The natural compounds which exhibited good ligand binding efficiency were selected and further subjected to pharmacokinetics and pharmacodynamic study. The natural compounds which exhibited good ligand binding efficiency were selected and further subjected to pharmacokinetics and pharmacodynamic study. The finalized complexes were selected for the simulation studies and MM/PBSA‐based binding free energy calculations. This study expands the possibilities for the development of shortlisted molecules as novel anti‐cancer compounds.

Investigation of the Cyclohexene Oxidation Mechanism Through the Direct Measurement of Organic Peroxy Radicals.

Monoterpenes, the second most abundant biogenic volatile organic compounds globally, are crucial in forming secondary organic aerosols, making their oxidation mechanisms vital for addressing climate change and air pollution. This study utilized cyclohexene as a surrogate to explore first-generation products from its ozonolysis through laboratory experiments and mechanistic modeling. We employed proton transfer reaction mass spectrometry with NH4+ ion sources (NH4+-CIMS) and a custom-built OH calibration source to quantify organic peroxy radicals (RO2) and closed-shell species. Under near-real atmospheric conditions in a Potential Aerosol Mass-Oxidation Flow Reactor, we identified 30 ozonolysis products, expanding previous data sets of low-oxygen compounds. Combined with simulations based on the Generator for Explicit Chemistry and Kinetics of Organics in the Atmosphere and relevant literature, our results revealed that OH dominates over ozone in cyclohexene oxidation at typical atmospheric oxidant levels with H-abstraction contributing 30% of initial RO2 radicals. Highly oxidized molecules primarily arise from RO2 autoxidation initiated by ozone, and at least 15% of ozone oxidation products follow the overlooked nonvinyl hydroperoxides pathway. Gaps remain especially in understanding RO2 cross-reactions, and the structural complexity of monoterpenes further complicates research. As emissions decrease and afforestation increases, understanding these mechanisms becomes increasingly critical.

Quantitative structure-property relationship of glass transition temperatures for organic compounds

The glass transition temperatures (T gs) of materials used in the manufacture of organic light-emitting diodes (OLEDs) determine their thermal stability. Three Dragon descriptors, TPC, RBF and TDB04s, were adopted to develop quantitative structure-property relationships (QSPR) for the prediction of the T gs of 66 compounds (Data Set I) for OLED application, by applying random forest (RF) and support vector machine (SVM). The RF Model A, based on a training set (44 compounds), was validated with a test set (22 compounds). The RF Model A possesses a coefficient of determination R 2 of 0.942 and a root mean square (rms) error of 10.750 K for the training set and R 2 of 0.909 and rms error of 11.102 K for the test set, which are more accurate than the results from the SVM model. The RF Model A was further validated with 63 OLED molecules (Data Set II). Moreover, the three Dragon descriptors (TPC, RBF and TDB04s) were used to build another T g QSPR model (named RF Model B) for a large dataset of 1934 OLED molecules (Data Set III), which achieved rms errors of 16.79 K for Data Set III, 22.89 K for Data Set I and 20.17 K for Data Set II.

Machine Learning and Deep Learning Models for Predicting Noncovalent Inhibitors of AmpC β-Lactamase.

Continuous use of antibiotics leads to the ability of bacteria to adapt by developing complex antibiotic resistance (AR) mechanisms. The synthesis of β-lactamases is a widely observed AR mechanism. The class C β-lactamase (AmpC) causes significant resistance toward β-lactam antibiotics, and new treatments are urgently needed. Noncovalent inhibitors have been developed against a broad spectrum of β-lactamases. In this study, we developed robust and accurate models for predicting noncovalent inhibitors of AmpC using large compound data sets and machine/deep learning modeling. We created support vector machine (SVM), random forest (RF), and feed-forward neural network (FFNN) classification models. The cross-validation (CV) accuracies varied between 80 and 82%, as combined models reached an accuracy of 83%. We analyzed the physicochemical characteristics of the noncovalent inhibitors and predicted the binding modes for some of them. Such models are helpful for identifying new noncovalent inhibitors in order to establish novel solutions against the growing resistance to standard β-lactam inhibitors. The best RF, SVM, and FFNN models for predicting noncovalent inhibitors of AmpC β-lactamase are available in the GitHub repository, https://github.com/UPCmctr/ML-DL-AmpC-B-lactamase.

The 3D-QSAR Study, Molecular Docking, and ADMET Analysis of Darunavir Derivatives of HIV-1 Protease Inhibitors

Background: Acquired Immunodeficiency Syndrome (AIDS) is one of most prevalent infectious diseases in the world , and HIV-1 protease (PR) is a vital target of drug design. Nowadays, three-dimensional quantitative structure-activity relationships (3D-QSAR) are applied to help design new protease inhibitions (PIs). Objective: The primary objective of this study is to apply the 3D-QSAR study to a series of 42 derivatives of Darunavir (DRV) and to design new molecules possessing high antivirus activity. Methods: Partial Least Squares (PLS) were used to cross-validate the dataset of compounds, and the optimal number of principal components (ONC), cross-validate coefficient (q²), standard error of estimate (SEE), non-cross-validated correlation coefficient (R²) and fisher test value (F) were calculated to assess model robustness. In this study, the CoMSIA-DAH model (q²=0.754, r²= 0.988, rpred2=0.57) possessed the highest predicted activity. Newly designed molecules were analyzed by docking studies with compound 25 taken as a template. Results: Within eight newly designed drugs, compound N02 possessed the highest antivirus activity (IC50=0.00461 nM) predicted by the CoMSIA-DAH model. The Surflex-Dock module of SYBYL-X 2.0 was used to affirm the predicted anti-PR activity of the newly designed compounds and the results of docking complex structure could be visualized. All newly designed molecules were in agreement with CSore above four and the docking study revealed that Asp29, Asp30, Ile50, Asp124, Asp128, Asp129 and Ile149 were critical residues in the process of inhibiting PR. Conclusion: One of the main aspects of this study is the successful design of a series of molecules with excellent investigatory values, which elucidates explicit quantitative structure-activity relationships of DRV derivatives and will provide significant suggestions for future pharmaceutical research.

Label distribution learning for compound facial expression recognition in‐the‐wild: A comparative study

AbstractHuman emotional states encompass both basic and compound facial expressions. However, current works primarily focus on basic expressions, consequently neglecting the broad spectrum of human emotions encountered in practical scenarios. Compound facial expressions involve the simultaneous manifestation of multiple emotions on an individual's face. This phenomenon reflects the complexity and richness of human states, where facial features dynamically convey a combination of feelings. This study embarks on a pioneering exploration of Compound Facial Expression Recognition (CFER), with a distinctive emphasis on leveraging the Label Distribution Learning (LDL) paradigm. This strategic application of LDL aims to address the ambiguity and complexity inherent in compound expressions, marking a significant departure from the dominant Single Label Learning (SLL) and Multi‐Label Learning (MLL) paradigms. Within this framework, we rigorously investigate the potential of LDL for a critical challenge in Facial Expression Recognition (FER): recognizing compound facial expressions in uncontrolled environments. We utilize the recently introduced RAF‐CE dataset, meticulously designed for compound expression assessment. By conducting a comprehensive comparative analysis pitting LDL against conventional SLL and MLL approaches on RAF‐CE, we aim to definitively establish LDL's superiority in handling this complex task. Furthermore, we assess the generalizability of LDL models trained on RAF‐CE by evaluating their performance on the EmotioNet and RAF‐DB Compound datasets. This demonstrates their effectiveness without domain adaptation. To solidify these findings, we conduct a comprehensive comparative analysis of 12 cutting‐edge LDL algorithms on RAF‐CE, S‐BU3DFE, and S‐JAFFE datasets, providing valuable insights into the most effective LDL techniques for FER in‐the‐wild.

Data-Driven Modelling of Substituted Pyrimidine and Uracil-Based Derivatives Validated with Newly Synthesized and Antiproliferative Evaluated Compounds.

The pyrimidine heterocycle plays an important role in anticancer research. In particular, the pyrimidine derivative families of uracil show promise as structural scaffolds relevant to cervical cancer. This group of chemicals lacks data-driven machine learning quantitative structure-activity relationships (QSARs) that allow for generalization and predictive capabilities in the search for new active compounds. To achieve this, a dataset of pyrimidine and uracil compounds from ChEMBL were collected and curated. A workflow was developed for data-driven machine learning QSAR using an intuitive dataset design and forwards selection of molecular descriptors. The model was thoroughly externally validated against available data. Blind validation was also performed by synthesis and antiproliferative evaluation of new synthesized uracil-based and pyrimidine derivatives. The most active compound among new synthesized derivatives, 2,4,5-trisubstituted pyrimidine was predicted with the QSAR model with differences of 0.02 compared to experimentally tested activity.

Neglected Tropical Diseases: A Chemoinformatics Approach for the Use of Biodiversity in Anti-Trypanosomatid Drug Discovery.

The development of new treatments for neglected tropical diseases (NTDs) remains a major challenge in the 21st century. In most cases, the available drugs are obsolete and have limitations in terms of efficacy and safety. The situation becomes even more complex when considering the low number of new chemical entities (NCEs) currently in use in advanced clinical trials for most of these diseases. Natural products (NPs) are valuable sources of hits and lead compounds with privileged scaffolds for the discovery of new bioactive molecules. Considering the relevance of biodiversity for drug discovery, a chemoinformatics analysis was conducted on a compound dataset of NPs with anti-trypanosomatid activity reported in 497 research articles from 2019 to 2024. Structures corresponding to different metabolic classes were identified, including terpenoids, benzoic acids, benzenoids, steroids, alkaloids, phenylpropanoids, peptides, flavonoids, polyketides, lignans, cytochalasins, and naphthoquinones. This unique collection of NPs occupies regions of the chemical space with drug-like properties that are relevant to anti-trypanosomatid drug discovery. The gathered information greatly enhanced our understanding of biologically relevant chemical classes, structural features, and physicochemical properties. These results can be useful in guiding future medicinal chemistry efforts for the development of NP-inspired NCEs to treat NTDs caused by trypanosomatid parasites.

Sulfotransferase-mediated phase II drug metabolism prediction of substrates and sites using accessibility and reactivity-based algorithms.

Sulphotransferases (SULTs) are a major phase II metabolic enzyme class contributing ~20 % to the Phase II metabolism of FDA-approved drugs. Ignoring the potential for SULT-mediated metabolism leaves a strong potential for drug-drug interactions, often causing late-stage drug discovery failures or black-boxed warnings on FDA labels. The existing models use only accessibility descriptors and machine learning (ML) methods for class and site of sulfonation (SOS) predictions for SULT. In this study, a variety of accessibility, reactivity, and hybrid models and algorithms have been developed to make accurate substrate and SOS predictions. Unlike the literature models, reactivity parameters for the aliphatic or aromatic hydroxyl groups (R/Ar-O-H), the Bond Dissociation Energy (BDE) gave accurate models with a True Positive Rate (TPR)=0.84 for SOS predictions. We offer mechanistic insights to explain these novel findings that are not recognized in the literature. The accessibility parameters like the ratio of Chemgauss4 Score (CGS) and Molecular Weight (MW) CGS/MW and distance from cofactor (Dis) were essential for class predictions and showed TPR=0.72. Substrates consistently had lower BDE, Dis, and CGS/MW than non-substrates. Hybrid models also performed acceptablely for SOS predictions. Using the best models, Algorithms gave an acceptable performance in class prediction: TPR=0.62, False Positive Rate (FPR)=0.24, Balanced accuracy (BA)=0.69, and SOS prediction: TPR=0.98, FPR=0.60, and BA=0.69. A rule-based method was added to improve the predictive performance, which improved the algorithm TPR, FPR, and BA. Validation using an external dataset of drug-like compounds gave class prediction: TPR=0.67, FPR=0.00, and SOS prediction: TPR=0.80 and FPR=0.44 for the best Algorithm. Comparisons with standard ML models also show that our algorithm shows higher predictive performance for classification on external datasets. Overall, these models and algorithms (SOS predictor) give accurate substrate class and site (SOS) predictions for SULT-mediated Phase II metabolism and will be valuable to the drug discovery community in academia and industry. The SOS predictor is freely available for academic/non-profit research via the GitHub link.

Cheminformatics analysis of indoleamine and tryptophan 2,3-dioxygenase inhibitors: A descriptor and fingerprint based machine learning approach to disclose selectivity measures

Indoleamine 2,3-dioxygenase (IDO) and tryptophan 2,3-dioxygenase (TDO) are attractive drug targets for cancer immunotherapy. After disappointing results of the epacadostat as a selective IDO inhibitor in phase III clinical trials, there is much interest in the development of the TDO selective inhibitors. In the current study, several data analysis methods and machine learning approaches including logistic regression, Random Forest, XGBoost and Support Vector Machines were used to model a data set of compounds retrieved from ChEMBL. Models based on the Morgan fingerprints revealed notable fragments for the selective inhibition of the IDO, TDO or both. Multiple fragment docking was performed to find the best set of bound fragments and their orientation in the space for efficient linking. Linking the fragments and optimization of the final molecules were accomplished by means of an artificial intelligence generative framework. Finally, selectivity of the optimized molecules was assessed and the top 4 lead molecules were filtered through PAINS, Brenk and NIH filters. Results indicated that phenyloxalamide, fluoroquinoline, and 3-bromo-4-fluroaniline confer selectivity towards the IDO inhibition. Correspondingly, 1-benzyl-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione was found to be an integral fragment for the selective inhibition of the TDO by constituting a coordination bond with the Fe atom of heme. In addition, furo[2,3-c]pyridine-2,3-diamine was found as a common fragment for inhibition of the both targets and can be used in the design of the dual target inhibitors of the IDO and TDO. The new fragments introduced here can be a useful building blocks for incorporation into the selective TDO or dual IDO/TDO inhibitors.

Integrated re-analysis of transcriptomic and proteomic datasets reveals potential mechanisms for Zika viral-based oncolytic therapy in neuroblastoma.

Paediatric neuroblastoma and brain tumours account for a third of all childhood cancer-related mortality. High-risk neuroblastoma is highly aggressive and survival is poor despite intensive multi-modal therapies with significant toxicity. Novel therapies are desperately needed. The Zika virus (ZIKV) can access the nervous system and there is growing interest in employing ZIKV as a potential therapy against paediatric nervous system tumours, including neuroblastoma. Here, we perform extensive data mining, integration and re-analysis of ZIKV infection datasets to highlight molecular mechanisms that may govern the oncolytic response in neuroblastoma cells. We collate infection data of multiple neuroblastoma cell lines by different ZIKV strains from a body of published literature to inform the susceptibility of neuroblastoma to the ZIKV oncolytic response. Integrating published transcriptomics, interaction proteomics, dependency factor and compound datasets we propose the involvement of multiple host systems during ZIKV infection. Through data mining of published literature, we observed most paediatric neuroblastoma cell lines to be highly susceptible to ZIKV infection and propose the PRVABC59 ZIKV strain to be the most promising candidate for neuroblastoma oncolytic virotherapy. ZIKV induces TNF signalling, lipid metabolism, the Unfolded Protein Response (UPR), and downregulates cell cycle and DNA replication processes. ZIKV infection is dependent on sterol regulatory element binding protein (SREBP)-regulated lipid metabolism and three protein complexes; V-ATPase, ER Membrane Protein Complex (EMC) and mammalian translocon. We propose ZIKV non-structural protein 4B (NS4B) as a likely mediator of ZIKVs interaction with IRE1-mediated UPR, lipid metabolism and mammalian translocon. Our work provides a significant understanding of ZIKV infection in neuroblastoma cells, which will facilitate the progression of ZIKV-based oncolytic virotherapy through pre-clinical research and clinical trials.

Predicting drug solubility in organic solvents mixtures: A machine-learning approach supported by high-throughput experimentation

A novel approach based on supervised machine-learning is proposed to predict the solubility of drugs and drug-like molecules in mixtures of organic solvents. Similar to quantitative structure–property relationship (QSPR) models, different solvent types are identified by molecular descriptors, which, in this study, are considered as UNIFAC subgroups. To overcome the potential lack of UNIFAC subgroups for the complex Active Pharmaceutical Ingredients (APIs) currently developed in the pharmaceutical industry, the API molecule is considered as a unique entity in the proposed modelling approach. Therefore, API solubility is predicted as a function of temperature, functional subgroups of the solvents and composition of the solvent mixture; in turn, regressors’ correlation is handled through Partial Least-Squares (PLS) regression. The method is developed and tested with experimental data of a real API and 14 organic solvents that are industrially employed for crystallisation. Solubility predictions are accurate and precise for single solvents, binary mixtures and ternary mixtures of organic solvents at different compositions and temperatures, with a determination coefficient R2 ≥ 0.90.To further test the applicability of the model, the proposed approach is applied to 9 literature organic solubility datasets of drugs and drug-like compounds and compared to benchmark solubility models in the literature. Results show that the proposed approach provides satisfactory predictions: the majority of validation and calibration data have R2 = 0.95–0.99; the ratio between RMSE (root mean squared error) of the proposed method and the range of measured solubility values is from 1 to 3 orders of magnitude smaller than the RMSE ratio obtained by the benchmark models.

Small-Molecule Inhibitors of TIPE3 Protein Identified through Deep Learning Suppress Cancer Cell Growth In Vitro.

Tumor necrosis factor-α-induced protein 8-like 3 (TNFAIP8L3 or TIPE3) functions as a transfer protein for lipid second messengers. TIPE3 is highly upregulated in several human cancers and has been established to significantly promote tumor cell proliferation, migration, and invasion and inhibit the apoptosis of cancer cells. Thus, inhibiting the function of TIPE3 is expected to be an effective strategy against cancer. The advancement of artificial intelligence (AI)-driven drug development has recently invigorated research in anti-cancer drug development. In this work, we incorporated DFCNN, Autodock Vina docking, DeepBindBC, MD, and metadynamics to efficiently identify inhibitors of TIPE3 from a ZINC compound dataset. Six potential candidates were selected for further experimental study to validate their anti-tumor activity. Among these, three small-molecule compounds (K784-8160, E745-0011, and 7238-1516) showed significant anti-tumor activity in vitro, leading to reduced tumor cell viability, proliferation, and migration and enhanced apoptotic tumor cell death. Notably, E745-0011 and 7238-1516 exhibited selective cytotoxicity toward tumor cells with high TIPE3 expression while having little or no effect on normal human cells or tumor cells with low TIPE3 expression. A molecular docking analysis further supported their interactions with TIPE3, highlighting hydrophobic interactions and their shared interaction residues and offering insights for designing more effective inhibitors. Taken together, this work demonstrates the feasibility of incorporating deep learning and MD simulations in virtual drug screening and provides inhibitors with significant potential for anti-cancer drug development against TIPE3-.

A Machine Learning Model for Evaluation of the Corrosion Inhibition Capacity of Quinoxaline Compounds

Investigating potential corrosion inhibitors via empirical research is a labor- and resource-intensive process. In this work, we evaluated various linear and non-linear algorithms as predictive models for corrosion inhibition efficiency (CIE) values using a machine learning (ML) paradigm based on the quantitative structure-property relationship (QSPR) model. In the quinoxaline compound dataset, our analysis showed that the XGBoost model performed the best predictor of other ensemble-based models. The coefficient of determination (R2), mean absolute percentage error (MAPE), and root mean squared error (RMSE) metrics were used to objectively assess this superiority. To sum up, our study offers a fresh viewpoint on the effectiveness of machine learning algorithms in determining the ability of organic compounds like quinoxaline to suppress corrosion on iron surfaces.