Molecular Descriptors

Physicochemical modelling of the retention mechanism of temperature-responsive polymeric columns for HPLC through machine learning algorithms

Temperature-responsive liquid chromatography (TRLC) offers a promising alternative to reversed-phase liquid chromatography (RPLC) for environmentally friendly analytical techniques by utilizing pure water as a mobile phase, eliminating the need for harmful organic solvents. TRLC columns, packed with temperature-responsive polymers coupled to silica particles, exhibit a unique retention mechanism influenced by temperature-induced polymer hydration. An investigation of the physicochemical parameters driving separation at high and low temperatures is crucial for better column manufacturing and selectivity control. Assessment of predictability using a dataset of 139 molecules analyzed at different temperatures elucidated the molecular descriptors (MDs) relevant to retention mechanisms. Linear regression, support vector regression (SVR), and tree-based ensemble models were evaluated, with no standout performer. The precision, accuracy, and robustness of models were validated through metrics, such as r and mean absolute error (MAE), and statistical analysis. At 45 ∘ C , logP predominantly influenced retention, akin to reversed-phase columns, while at5 ∘ C , complex interactions with lipophilic and negative MDs, along with specific functional groups, dictated retention. These findings provide deeper insights into TRLC mechanisms, facilitating method development and maximizing column potential.

Predictive binary mixture toxicity modeling of fluoroquinolones (FQs) and the projection of toxicity of hypothetical binary FQ mixtures: a combination of 2D-QSAR and machine-learning approaches

All sorts of chemicals get degraded under various environmental stresses, and the degradates coexist with the parent compounds as mixtures in the environment. Antibiotics emerge as an additional concern due to the bioactive nature of both the parent compound and degradation products and their combined exposure to the environment. Therefore, environmental risk assessment of antibiotics and their degradation products is very much necessary. In this direction, we made use of in silico new approach methodologies (NAMs) and machine-learning algorithms. In this study, we have developed a robust and predictive mixture-quantitative structure-activity relationship (QSAR) model with promising quality and predictability (internal: MAETrain = 0.085, QLOO2 = 0.849, external: MAETest = 0.090, and QF12 = 0.859) for predicting the toxicity of the mixtures of a class of antibiotics and their degradation products. To obtain the predictive model, toxicity data of 78 binary fluoroquinolone mixtures in E. coli (endpoint: log 1/IC50 in molar) have been utilized. We have used only 0D-2D descriptors to efficiently encode the structural features of mixture components without any additional complexities. The optimization of the class of mixture descriptors has been performed in this study by using three different mixing rules (linear combination of molecular contributions, the squared molecular contributions, and the norm of molecular contributions). Different machine-learning approaches namely, random forest (RF), ada boost, gradient boost (GB), extreme gradient boost (XGB), support vector machine (SVM), linear support vector machine (LSVM), and ridge regression (RR) have been employed here apart from the conventional partial least squares (PLS) regression to optimize the modeling approach. A rigorous validation protocol has been used for assessing the goodness-of-fit, robustness, and external predictability of the models. Finally, the toxicity of possible untested mixtures of different photodegradation products of fluoroquinolones has been predicted using the best model reported in this study.

Predicting Free Energies of Exfoliation and Solvation for Graphitic Carbon Nitrides Using Machine Learning

As a metal-free and visible-light-responsive photocatalyst, graphitic carbon nitride (g-C3N4) has emerged as a new research hotspot and has attracted broad attention in the field of solar energy conversion and thin-film transistors. Liquid-phase exfoliation (LPE) is the best-known method for the synthesis of 2D g-C3N4 nanosheets. In LPE, bulk g-C3N4 is exfoliated in a solvent via high-shear mixing or sonication in order to produce a stable suspension of individual nanosheets. Two parameters of importance in gauging the performance of a solvent in LPE are the free energy required to exfoliate a unit area of layered materials into individual sheets in the solvent (ΔGexf) and the solvation free energy per unit area of a nanosheet (ΔGsol). While approximations for the free energies exist, they are shown in our previous work to be inaccurate and incapable of capturing the experimentally observed efficacy of LPE. Molecular dynamics (MD) simulations can provide accurate free-energy calculations, but doing so for every single solvent is time- and resource-consuming. Herein, machine learning (ML) algorithms are used to predict ΔGexf and ΔGsol for g-C3N4. First, a database for ΔGexf and ΔGsol is created based on a series of MD simulations involving 49 different solvents with distinct chemical structures and properties. The data set also includes values of critical descriptors for the solvents, including density, surface tension, dielectric constant, etc. Different ML methods are compared, accompanied by descriptor selection, to develop the most accurate model for predicting ΔGexf and ΔGsol. The extra tree regressor is shown to be the best performer among the six ML methods studied. Experimental validation of the model is conducted by performing dispersibility tests in several solvents for which the free energies are predicted. Finally, the influence of the selected descriptors on the free energies is analyzed, and strategies for solvent selection in LPE are proposed.

Artificial intelligence for natural product drug discovery

Developments in computational omics technologies have provided new means to access the hidden diversity of natural products, unearthing new potential for drug discovery. In parallel, artificial intelligence approaches such as machine learning have led to exciting developments in the computational drug design field, facilitating biological activity prediction and de novo drug design for molecular targets of interest. Here, we describe current and future synergies between these developments to effectively identify drug candidates from the plethora of molecules produced by nature. We also discuss how to address key challenges in realizing the potential of these synergies, such as the need for high-quality datasets to train deep learning algorithms and appropriate strategies for algorithm validation.

APDB: a database on air pollutant characterization and similarity prediction

The World Health Organization estimates that 9 out of 10 people worldwide breathe air containing high levels of pollutants. Long-term and chronic exposure to high concentrations of air pollutants is associated with deleterious effects on vital organs, including increased inflammation in the lungs, oxidative stress in the heart and disruption of the blood-brain barrier. For this reason, in an effort to find an association between exposure to pollutants and the toxicological effects observable on human health, an online resource collecting and characterizing in detail pollutant molecules could be helpful to investigate their properties and mechanisms of action. We developed a database, APDB, collecting air-pollutant-related data from different online resources, in particular, molecules from the US Environmental Protection Agency, their associated targets and bioassays found in the PubChem chemical repository and their computed molecular descriptors and quantum mechanics properties. A web interface allows (i) to browse data by category, (ii) to navigate the database by querying molecules and targets and (iii) to visualize and download molecule and target structures as well as computed descriptors and similarities. The desired data can be freely exported in textual/tabular format and the whole database in SQL format. Database URL http://apdb.di.univr.it.

Large-scale evaluation of k-fold cross-validation ensembles for uncertainty estimation

It is insightful to report an estimator that describes how certain a model is in a prediction, additionally to the prediction alone. For regression tasks, most approaches implement a variation of the ensemble method, apart from few exceptions. Instead of a single estimator, a group of estimators yields several predictions for an input. The uncertainty can then be quantified by measuring the disagreement between the predictions, for example by the standard deviation. In theory, ensembles should not only provide uncertainties, they also boost the predictive performance by reducing errors arising from variance. Despite the development of novel methods, they are still considered the "golden-standard" to quantify the uncertainty of regression models. Subsampling-based methods to obtain ensembles can be applied to all models, regardless whether they are related to deep learning or traditional machine learning. However, little attention has been given to the question whether the ensemble method is applicable to virtually all scenarios occurring in the field of cheminformatics. In a widespread and diversified attempt, ensembles are evaluated for 32 datasets of different sizes and modeling difficulty, ranging from physicochemical properties to biological activities. For increasing ensemble sizes with up to 200 members, the predictive performance as well as the applicability as uncertainty estimator are shown for all combinations of five modeling techniques and four molecular featurizations. Useful recommendations were derived for practitioners regarding the success and minimum size of ensembles, depending on whether predictive performance or uncertainty quantification is of more importance for the task at hand.

How the Structure of Per- and Polyfluoroalkyl Substances (PFAS) Influences Their Binding Potency to the Peroxisome Proliferator-Activated and Thyroid Hormone Receptors-An In Silico Screening Study

In this study, we investigated PFAS (per- and polyfluoroalkyl substances) binding potencies to nuclear hormone receptors (NHRs): peroxisome proliferator-activated receptors (PPARs) α, β, and γ and thyroid hormone receptors (TRs) α and β. We have simulated the docking scores of 43 perfluoroalkyl compounds and based on these data developed QSAR (Quantitative Structure-Activity Relationship) models for predicting the binding probability to five receptors. In the next step, we implemented the developed QSAR models for the screening approach of a large group of compounds (4464) from the NORMAN Database. The in silico analyses indicated that the probability of PFAS binding to the receptors depends on the chain length, the number of fluorine atoms, and the number of branches in the molecule. According to the findings, the considered PFAS group bind to the PPARα, β, and γ only with low or moderate probability, while in the case of TR α and β it is similar except that those chemicals with longer chains show a moderately high probability of binding.

An Optimized Cloud Computing Method for Extracting Molecular Descriptors

Extracting molecular descriptors from chemical compounds is an essential preprocessing phase for developing accurate classification models. Supervised machine learning algorithms offer the capability to detect "hidden" patterns that may exist in a large dataset of compounds, which are represented by their molecular descriptors. Assuming that molecules with similar structure tend to share similar physicochemical properties, large chemical libraries can be screened by applying similarity sourcing techniques in order to detect potential bioactive compounds against a molecular target. However, the process of generating these compound features is time-consuming. Our proposed methodology not only employs cloud computing to accelerate the process of extracting molecular descriptors but also introduces an optimized approach to utilize the computational resources in the most efficient way.

In silico prediction of siRNA ionizable-lipid nanoparticles In vivo efficacy: Machine learning modeling based on formulation and molecular descriptors

In silico prediction of the in vivo efficacy of siRNA ionizable-lipid nanoparticles is desirable as it can save time and resources dedicated to wet-lab experimentation. This study aims to computationally predict siRNA nanoparticles in vivo efficacy. A data set containing 120 entries was prepared by combining molecular descriptors of the ionizable lipids together with two nanoparticles formulation characteristics. Input descriptor combinations were selected by an evolutionary algorithm. Artificial neural networks, support vector machines and partial least squares regression were used for QSAR modeling. Depending on how the data set is split, two training sets and two external validation sets were prepared. Training and validation sets contained 90 and 30 entries respectively. The results showed the successful predictions of validation set log (siRNA dose) with R_val ²= 0.86-0.89 and 0.75-80 for validation sets one and two, respectively. Artificial neural networks resulted in the best R_val ² for both validation sets. For predictions that have high bias, improvement of R_val ² from 0.47 to 0.96 was achieved by selecting the training set lipids lying within the applicability domain. In conclusion, in vivo performance of siRNA nanoparticles was successfully predicted by combining cheminformatics with machine learning techniques.

The Alleviation of Dextran Sulfate Sodium (DSS)-Induced Colitis Correlate with the logP Values of Food-Derived Electrophilic Compounds

Food-derived electrophilic compounds (FECs) are small molecules with electrophilic groups with potential cytoprotective effects. This study investigated the differential effects of six prevalent FECs on colitis in dextran sodium sulfate (DSS)-induced mice and the underlying relationship with molecular characteristics. Fumaric acid (FMA), isoliquiritigenin (ISO), cinnamaldehyde (CA), ferulic acid (FA), sulforaphane (SFN), and chlorogenic acid (CGA) exhibited varying improvements in colitis on clinical signs, colonic histopathology, inflammatory and oxidative indicators, and Nrf2 pathway in a sequence of SFN, ISO > FA, CA > FMA, CGA. Representative molecular characteristics of the “penetration-affinity−covalent binding” procedure, logP value, Keap1 affinity energy, and electrophilic index of FECs were theoretically calculated, among which logP value revealed a strong correlation with colitis improvements, which was related to the expression of Nrf2 and its downstream proteins. Above all, SFN and ISO possessed high logP values and effectively improving DSS-induced colitis by activating the Keap1−Nrf2 pathway to alleviate oxidative stress and inflammatory responses.

Noise-robust optimization of quantum machine learning models for polymer properties using a simulator and validated on the IonQ quantum computer

Quantum machine learning for predicting the physical properties of polymer materials based on the molecular descriptors of monomers was investigated. Under the stochastic variation of the expected predicted values obtained from quantum circuits due to finite sampling, the methods proposed in previous works did not make sufficient progress in optimizing the parameters. To enable parameter optimization despite the presence of stochastic variations in the expected values, quantum circuits that improve prediction accuracy without increasing the number of parameters and parameter optimization methods that are robust to stochastic variations in the expected predicted values, were investigated. The multi-scale entanglement renormalization ansatz circuit improved the prediction accuracy without increasing the number of parameters. The stochastic gradient descent method using the parameter-shift rule for gradient calculation was shown to be robust to sampling variability in the expected value. Finally, the quantum machine learning model was trained on an actual ion-trap quantum computer. At each optimization step, the coefficient of determination \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^{2}$$\end{document}R2 improved equally on the actual machine and simulator, indicating that our findings enable the training of quantum circuits on the actual quantum computer to the same extent as on the simulator.

Extended Connectivity Fingerprints as a Chemical Reaction Representation for Enantioselective Organophosphorus-Catalyzed Asymmetric Reaction Prediction

Predicting the outcomes of organic reactions using data-driven approaches aids in the acceleration of research. In laboratory-scale experiments, only a small number of reaction data can be accessed for machine learning model construction, where reaction representations play a pivotal role in the success of model construction. Nevertheless, representation comparison for a small data set is not adequate. Herein, focusing on the enantioselectivity of phosphoric-acid-catalyzed reactions, various two-dimensional and three-dimensional reaction representations (descriptors) were compared. Overall, the concatenated form of the extended connectivity fingerprints showed the best predictive capability for the two types of data sets: high-throughput experimental data and manually collected literature data sets. Furthermore, highlighting the substructure contribution to the prediction outcome was shown to be informative for guiding catalyst development.

Prediction of Combined Sorbent and Catalyst Materials for SE-SMR, Using QSPR and Multitask Learning

The process of sorption enhanced steam methane reforming (SE-SMR) is an emerging technology for the production of low carbon hydrogen. The development of a suitable catalytic material, as well as a CO₂ adsorbent with high capture capacity, has slowed the upscaling of this process to date. In this study, to aid the development of a combined sorbent catalyst material (CSCM) for SE-SMR, a novel approach involving quantitative structure–property relationship analysis (QSPR) has been proposed. Through data-mining, two databases have been developed for the prediction of the last cycle capacity (g_CO2/g_sorbent) and methane conversion (%). Multitask learning (MTL) was applied for the prediction of CSCM properties. Patterns in the data of this study have also yielded further insights; colored scatter plots were able to show certain patterns in the input data, as well as suggestions on how to develop an optimal material. With the results from the actual vs predicted plots collated, raw materials and synthesis conditions were proposed that could lead to the development of a CSCM that has good performance with respect to both the last cycle capacity and the methane conversion.

Prediction reliability of QSAR models: an overview of various validation tools

The reliability of any quantitative structure-activity relationship (QSAR) model depends on multiple aspects such as the accuracy of the input dataset, selection of significant descriptors, the appropriate splitting process of the dataset, statistical tools used, and most notably on the measures of validation. Validation, the most crucial step in QSAR model development, confirms the reliability of the developed QSAR models and the acceptability of each step in the model development. The present review deals with various validation tools that involve multiple techniques that improve the model quality and robustness. The double cross-validation tool helps in building improved quality models using different combinations of the same training set in an inner cross-validation loop. This exhaustive method is also integrated for small datasets (< 40 compounds) in another tool, namely the small dataset modeler tool. The main aim of QSAR researchers is to improve prediction quality by lowering the prediction errors for the query compounds. 'Intelligent' selection of multiple models and consensus predictions integrated in the intelligent consensus predictor tool were found to be more externally predictive than individual models. Furthermore, another tool called Prediction Reliability Indicator was explained to understand the quality of predictions for a true external set. This tool uses a composite scoring technique to identify query compounds as 'good' or 'moderate' or 'bad' predictions. We have also discussed a quantitative read-across tool which predicts a chemical response based on the similarity with structural analogues. The discussed tools are freely available from https://dtclab.webs.com/software-tools or http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/ and https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home (for read-across).

Artificial intelligence: machine learning for chemical sciences

Research in molecular sciences witnessed the rise and fall of Artificial Intelligence (AI)/ Machine Learning (ML) methods, especially artificial neural networks, few decades ago. However, we see a major resurgence in the use of modern ML methods in scientific research during the last few years. These methods have had phenomenal success in the areas of computer vision, speech recognition, natural language processing (NLP), etc. This has inspired chemists and biologists to apply these algorithms to problems in natural sciences. Availability of high performance Graphics Processing Unit (GPU) accelerators, large datasets, new algorithms, and libraries has enabled this surge. ML algorithms have successfully been applied to various domains in molecular sciences by providing much faster and sometimes more accurate solutions compared to traditional methods like Quantum Mechanical (QM) calculations, Density Functional Theory (DFT) or Molecular Mechanics (MM) based methods, etc. Some of the areas where the potential of ML methods are shown to be effective are in drug design, prediction of high-level quantum mechanical energies, molecular design, molecular dynamics materials, and retrosynthesis of organic compounds, etc. This article intends to conceptually introduce various modern ML methods and their relevance and applications in computational natural sciences. Synopsis Recent surge in the application of machine learning (ML) methods in fundamental sciences has led to a perspective that these methods may become important tools in chemical science. This perspective provides an overview of the modern ML methods and their successful applications in chemistry during the last few years.

A Classification Model to Identify Direct-Acting Mutagenic Polycyclic Aromatic Hydrocarbon Transformation Products

Polycyclic aromatic hydrocarbons (PAHs) are a complex group of environmental contaminants, many having long environmental half-lives. As these compounds degrade, the changes in their structure can result in a substantial increase in mutagenicity compared to the parent compound. Over time, each individual PAH can potentially degrade into several thousand unique transformation products, creating a complex, constantly evolving set of intermediates. Microbial degradation is the primary mechanism of their transformation and ultimate removal from the environment, and this process can result in mutagenic activation similar to the metabolic activation that can occur in multicellular organisms. The diversity of the potential intermediate structures in PAH-contaminated environments renders hazard assessment difficult for both remediation professionals and regulators. A mixture of structural and energetic descriptors has proven effective in existing studies for classifying which PAH transformation products will be mutagenic. However, most existing studies of environmental PAH mutagens primarily focus on nitrogenated derivatives, which are prevalent in the atmosphere and not as relevant in soil. Additionally, PAH products commonly found in the environment can range from as large as five rings to as small as a single ring, requiring a broadly inclusive methodology to comprehensively evaluate mutagenic potential. We developed a combination of supervised and unsupervised machine learning methods to predict environmentally induced PAH mutagenicity with improved performance over currently available tools. K-means clustering with principal component analysis allows us to identify molecular clusters that we hypothesize to have similar mechanisms of action. Recursive feature elimination identifies the most influential descriptors. The cluster-specific regression outperforms available classifiers in predicting direct-acting mutagens resulting from the microbial biodegradation of PAHs and provides direction for future studies evaluating the environmental hazards resulting from PAH biodegradation.

MG-BERT: leveraging unsupervised atomic representation learning for molecular property prediction

MOTIVATION

Accurate and efficient prediction of molecular properties is one of the fundamental issues in drug design and discovery pipelines. Traditional feature engineering-based approaches require extensive expertise in the feature design and selection process. With the development of artificial intelligence (AI) technologies, data-driven methods exhibit unparalleled advantages over the feature engineering-based methods in various domains. Nevertheless, when applied to molecular property prediction, AI models usually suffer from the scarcity of labeled data and show poor generalization ability.

RESULTS

In this study, we proposed molecular graph BERT (MG-BERT), which integrates the local message passing mechanism of graph neural networks (GNNs) into the powerful BERT model to facilitate learning from molecular graphs. Furthermore, an effective self-supervised learning strategy named masked atoms prediction was proposed to pretrain the MG-BERT model on a large amount of unlabeled data to mine context information in molecules. We found the MG-BERT model can generate context-sensitive atomic representations after pretraining and transfer the learned knowledge to the prediction of a variety of molecular properties. The experimental results show that the pretrained MG-BERT model with a little extra fine-tuning can consistently outperform the state-of-the-art methods on all 11 ADMET datasets. Moreover, the MG-BERT model leverages attention mechanisms to focus on atomic features essential to the target property, providing excellent interpretability for the trained model. The MG-BERT model does not require any hand-crafted feature as input and is more reliable due to its excellent interpretability, providing a novel framework to develop state-of-the-art models for a wide range of drug discovery tasks.

First report on chemometric modeling of hydrolysis half-lives of organic chemicals

Hydrolysis is one of the most important processes of transformation of organic chemicals in water. The rates of reactions, final chemical entities of these processes, and half-lives of organic chemicals are of considerable interest to environmental chemists as well as authorities involved in the controlling the processing and disposal of such organic chemicals. In this study, we have proposed QSPR models for the prediction of hydrolysis half-life of organic chemicals as a function of different pH and temperature conditions using only two-dimensional molecular descriptors with definite physicochemical significance. For each model, suitable subsets of variables were elected using a genetic algorithm method; next, the elected subsets of variables were subjected to the best subset selection with a key objective to determine the best combination of descriptors for model generation. Finally, QSPR models were constructed using the best combination of variables employing the partial least squares (PLS) regression technique. Next, every final model was subjected for strict validation employing the internationally accepted internal and external validation parameters. The proposed models could be applicable for data gap filling to determine hydrolysis half-lives of organic chemicals at different environmental conditions. Generally, presence of aliphatic ether and ether functional groups, high percentage of oxygen content in the molecule and presence of O-Si pairs of atoms at topological distance one, results in a shorter hydrolysis half-life of organic chemicals. On the other hand, higher unsaturation content and high percentage of nitrogen content in molecules lead to higher hydrolysis half-life. It is also found that branched and compact molecules will have a lower half-life while straight chain analogues will have a higher half-life. To the best of our knowledge, the presented models are the first reported QSPR models for hydrolysis half-lives of organic chemicals at different pH values.

Identify Compounds' Target Against Alzheimer's Disease Based on In-Silico Approach

BACKGROUND

Alzheimer's disease swept every corner of the globe and the number of patients worldwide has been rising. At present, there are as many as 30 million people with Alzheimer's disease in the world, and it is expected to exceed 80 million people by 2050. Consequently, the study of Alzheimer's drugs has become one of the most popular medical topics.

METHODS

In this study, in order to build a predicting model for Alzheimer's drugs and targets, the attribute discriminators CfsSubsetEval, ConsistencySubsetEval and FilteredSubsetEval are combined with search methods such as BestFirst, GeneticSearch and Greedystepwise to filter the molecular descriptors. Then the machine learning algorithms such as BayesNet, SVM, KNN and C4.5 are used to construct the 2D-Structure Activity Relationship(2D-SAR) model. Its modeling results are utilized for Receiver Operating Characteristic curve(ROC) analysis.

RESULTS

The prediction rates of correctness using Randomforest for AChE, BChE, MAO-B, BACE1, Tau protein and Non-inhibitor are 77.0%, 79.1%, 100.0%, 94.2%, 93.2% and 94.9%, respectively, which are overwhelming as compared to those of BayesNet, BP, SVM, KNN, AdaBoost and C4.5.

CONCLUSION

In this paper, we conclude that Random Forest is the best learner model for the prediction of Alzheimer's drugs and targets. Besides, we set up an online server to predict whether a small molecule is the inhibitor of Alzheimer's target at http://47.106.158.30:8080/AD/. Furthermore, it can distinguish the target protein of a small molecule.

Prediction of K562 Cells Functional Inhibitors Based on Machine Learning Approaches

BACKGROUND

β thalassemia is a common monogenic genetic disease that is very harmful to human health. The disease arises is due to the deletion of or defects in β-globin, which reduces synthesis of the β-globin chain, resulting in a relatively excess number of α-chains. The formation of inclusion bodies deposited on the cell membrane causes a decrease in the ability of red blood cells to deform and a group of hereditary haemolytic diseases caused by massive destruction in the spleen.

METHODS

In this work, machine learning algorithms were employed to build a prediction model for inhibitors against K562 based on 117 inhibitors and 190 non-inhibitors.

RESULTS

The overall accuracy (ACC) of a 10-fold cross-validation test and an independent set test using Adaboost were 83.1% and 78.0%, respectively, surpassing Bayes Net, Random Forest, Random Tree, C4.5, SVM, KNN and Bagging.

CONCLUSION

This study indicated that Adaboost could be applied to build a learning model in the prediction of inhibitors against K526 cells.

Causation or only correlation? Application of causal inference graphs for evaluating causality in nano-QSAR models

In this paper, we suggest that causal inference methods could be efficiently used in Quantitative Structure-Activity Relationships (QSAR) modeling as additional validation criteria within quality evaluation of the model. Verification of the relationships between descriptors and toxicity or other activity in the QSAR model has a vital role in understanding the mechanisms of action. The well-known phrase "correlation does not imply causation" reflects insight statistically correlated with the endpoint descriptor may not cause the emergence of this endpoint. Hence, paradigmatic shifts must be undertaken when moving from traditional statistical correlation analysis to causal analysis of multivariate data. Methods of causal discovery have been applied for broader physical insight into mechanisms of action and interpretation of the developed nano-QSAR models. Previously developed nano-QSAR models for toxicity of 17 nano-sized metal oxides towards E. coli bacteria have been validated by means of the causality criteria. Using the descriptors confirmed by the causal technique, we have developed new models consistent with the straightforward causal-reasoning account. It was proven that causal inference methods are able to provide a more robust mechanistic interpretation of the developed nano-QSAR models.

Building a virtual ligand screening pipeline using free software: a survey

Virtual screening, the search for bioactive compounds via computational methods, provides a wide range of opportunities to speed up drug development and reduce the associated risks and costs. While virtual screening is already a standard practice in pharmaceutical companies, its applications in preclinical academic research still remain under-exploited, in spite of an increasing availability of dedicated free databases and software tools. In this survey, an overview of recent developments in this field is presented, focusing on free software and data repositories for screening as alternatives to their commercial counterparts, and outlining how available resources can be interlinked into a comprehensive virtual screening pipeline using typical academic computing facilities. Finally, to facilitate the set-up of corresponding pipelines, a downloadable software system is provided, using platform virtualization to integrate pre-installed screening tools and scripts for reproducible application across different operating systems.

Machine Learning Strategy for Accelerated Design of Polymer Dielectrics

The ability to efficiently design new and advanced dielectric polymers is hampered by the lack of sufficient, reliable data on wide polymer chemical spaces, and the difficulty of generating such data given time and computational/experimental constraints. Here, we address the issue of accelerating polymer dielectrics design by extracting learning models from data generated by accurate state-of-the-art first principles computations for polymers occupying an important part of the chemical subspace. The polymers are 'fingerprinted' as simple, easily attainable numerical representations, which are mapped to the properties of interest using a machine learning algorithm to develop an on-demand property prediction model. Further, a genetic algorithm is utilised to optimise polymer constituent blocks in an evolutionary manner, thus directly leading to the design of polymers with given target properties. While this philosophy of learning to make instant predictions and design is demonstrated here for the example of polymer dielectrics, it is equally applicable to other classes of materials as well.

Prediction of the Fate of Organic Compounds in the Environment From Their Molecular Properties: A Review

A comprehensive review of quantitative structure-activity relationships (QSAR) allowing the prediction of the fate of organic compounds in the environment from their molecular properties was done. The considered processes were water dissolution, dissociation, volatilization, retention on soils and sediments (mainly adsorption and desorption), degradation (biotic and abiotic), and absorption by plants. A total of 790 equations involving 686 structural molecular descriptors are reported to estimate 90 environmental parameters related to these processes. A significant number of equations was found for dissociation process (pKa), water dissolution or hydrophobic behavior (especially through the KOW parameter), adsorption to soils and biodegradation. A lack of QSAR was observed to estimate desorption or potential of transfer to water. Among the 686 molecular descriptors, five were found to be dominant in the 790 collected equations and the most generic ones: four quantum-chemical descriptors, the energy of the highest occupied molecular orbital (EHOMO) and the energy of the lowest unoccupied molecular orbital (ELUMO), polarizability (α) and dipole moment (μ), and one constitutional descriptor, the molecular weight. Keeping in mind that the combination of descriptors belonging to different categories (constitutional, topological, quantum-chemical) led to improve QSAR performances, these descriptors should be considered for the development of new QSAR, for further predictions of environmental parameters. This review also allows finding of the relevant QSAR equations to predict the fate of a wide diversity of compounds in the environment.

The effect of various atomic partial charge schemes to elucidate consensus activity-correlating molecular regions: a test case of diverse QSAR models

The estimation of atomic partial charges of the small molecules to calculate molecular interaction fields (MIFs) is an important process in field-based quantitative structure-activity relationship (QSAR). Several studies showed the influence of partial charge schemes that drastically affects the prediction accuracy of the QSAR model and focused on the selection of appropriate charge models that provide highest cross-validated correlation coefficient ([Formula: see text] or q(2)) to explain the variation in chemical structures against biological endpoints. This study shift this focus in a direction to understand the molecular regions deemed to explain SAR in various charge models and recognize a consensus picture of activity-correlating molecular regions. We selected eleven diverse dataset and developed MIF-based QSAR models using various charge schemes including Gasteiger-Marsili, Del Re, Merck Molecular Force Field, Hückel, Gasteiger-Hückel, and Pullman. The generalized resultant QSAR models were then compared with Open3DQSAR model to interpret the MIF descriptors decisively. We suggest the regions of activity contribution or optimization can be effectively determined by studying various charge-based models to understand SAR precisely.

Two new atom centered fragment descriptors and scoring function enhance classification of antibacterial activity

Classification of pharmacologic activity of a chemical compound is an essential step in any drug discovery process. We develop two new atom-centered fragment descriptors (vertex indices)--one based solely on topological considerations without discriminating atom or bond types, and another based on topological and electronic features. We also assess their usefulness by devising a method to rank and classify molecules with regard to their antibacterial activity. Classification performances of our method are found to be superior compared to two previous studies on large heterogeneous data sets for hit finding and hit-to-lead studies even though we use much fewer parameters. It is found that for hit finding studies topological features (simple graph) alone provide significant discriminating power, and for hit-to-lead process small but consistent improvement can be made by additionally including electronic features (colored graph). Our approach is simple, interpretable, and suitable for design of molecules as we do not use any physicochemical properties. The singular use of vertex index as descriptor, novel range based feature extraction, and rigorous statistical validation are the key elements of this study.

Sharp bounds and normalization of Wiener-type indices

Complex networks abound in physical, biological and social sciences. Quantifying a network’s topological structure facilitates network exploration and analysis, and network comparison, clustering and classification. A number of Wiener type indices have recently been incorporated as distance-based descriptors of complex networks, such as the R package QuACN. Wiener type indices are known to depend both on the network’s number of nodes and topology. To apply these indices to measure similarity of networks of different numbers of nodes, normalization of these indices is needed to correct the effect of the number of nodes in a network. This paper aims to fill this gap. Moreover, we introduce an -Wiener index of network , denoted by . This notion generalizes the Wiener index to a very wide class of Wiener type indices including all known Wiener type indices. We identify the maximum and minimum of over a set of networks with nodes. We then introduce our normalized-version of -Wiener index. The normalized -Wiener indices were demonstrated, in a number of experiments, to improve significantly the hierarchical clustering over the non-normalized counterparts.

Comparison of representational spaces based on structural information in the development of QSAR models for benzylamino enaminone derivatives

In this paper we study different representational spaces of molecule data sets based on 2D representation models for the building of QSAR models for the prediction of the activity of 37 benzylamino enaminone derivatives. Approximations based on classical similarity calculated from fingerprint representation of molecules and isomorphism obtained using sub-graph matching algorithms are compared to fragmentation-based approximations using partial least squares and genetic algorithms. The influence of the anchored position of a non-common moiety and the kind of substituents in the common core structure of the data set are analysed, demonstrating the anomalous behaviour of some molecules and therefore the difficulty in building prediction models. These problems are solved by considering approximate similarity models. These models tune the prediction equations based on the size of the substituent and the anchored position, by adjusting the contribution of each substituent in similarity measurements calculated between the molecule data sets.

New topological indices with very high discriminatory power

Several molecular descriptors are used in developing quantitative structure-activity relationships (QSARs). A large number of them are already in use and new descriptors are added every year. Two new topological indices with very high discriminatory power are reported in this paper. The two indices ranked all planar graphs of alkanes C(4) to C(6) uniquely and were found to have non-degenerate values for all the 7668 constitutional isomers (alkane trees) from C(4) to C(15). Low intercorrelation with several of the commonly used topological indices was studied using a diverse data set of 820 chemicals and the new indices proposed in the study were found to cluster in different nodes. This further confirmed their low intercorrelation with other molecular descriptors used in the study. The new descriptors were found to be useful in QSAR modelling.