Chemical Structure Information Research Articles

A novel TiC-TiN based spectrally selective absorbing coating: Structure, optical properties and thermal stability

A novel TiC-TiN/Al2O3 tandem solar selective absorbing coating was successfully deposited on Si (1 0 0) and stainless steel (SS) substrates by reactive RF magnetron sputtering deposition. The spectral characteristics, microstructure, thermal stability and solar-thermal conversion efficiency of the coating before and after a long annealing in vacuum were investigated systematically. The optical constants of the coating are measured by an Ultraviolet–visible-near-infrared (UV/Vis/NIR) spectrometer, and the chemical composition analysis, structure information and surface morphology are investigated using XRD, XPS and FESEM. An excellent absorptance (α = 0.92) and a low emittance (ε = 0.11) at 82 °C are obtained by the as-prepared coatings. After annealed at 500 °C for 100 h in vacuum, there is no significant change in micromorphology and optical properties of the coating, which produces a good solar-thermal conversion efficiency. These results indicate that it can be applied as a potential coating material to a high temperature concentrated solar power (CSP) system.

Using human in vitro transcriptome analysis to build trustworthy machine learning models for prediction of animal drug toxicity

During the development of new drugs or compounds there is a requirement for preclinical trials, commonly involving animal tests, to ascertain the safety of the compound prior to human trials. Machine learning techniques could provide an in-silico alternative to animal models for assessing drug toxicity, thus reducing expensive and invasive animal testing during clinical trials, for drugs that are most likely to fail safety tests. Here we present a machine learning model to predict kidney dysfunction, as a proxy for drug induced renal toxicity, in rats. To achieve this, we use inexpensive transcriptomic profiles derived from human cell lines after chemical compound treatment to train our models combined with compound chemical structure information. Genomics data due to its sparse, high-dimensional and noisy nature presents significant challenges in building trustworthy and transparent machine learning models. Here we address these issues by judiciously building feature sets from heterogenous sources and coupling them with measures of model uncertainty achieved through Gaussian Process based Bayesian models. We combine the use of insight into the feature-wise contributions to our predictions with the use of predictive uncertainties recovered from the Gaussian Process to improve the transparency and trustworthiness of the model.

Direct Comparison of Total Clearance Prediction: Computational Machine Learning Model versus Bottom-Up Approach Using In Vitro Assay

The in vitro-in vivo extrapolation (IVIVE) approach for predicting total plasma clearance (CLtot) has been widely used to rank order compounds early in discovery. More recently, a computational machine learning approach utilizing physicochemical descriptors and fingerprints calculated from chemical structure information has emerged, enabling virtual predictions even earlier in discovery. Previously, this approach focused more on in vitro intrinsic clearance (CLint) prediction. Herein, we directly compare these two approaches for predicting CLtot in rats. A structurally diverse set of 1114 compounds with known in vivo CLtot, in vitro CLint, and plasma protein binding was used as the basis for this evaluation. The machine learning models were assessed by validation approaches using the time- and cluster-split training and test sets, and five-fold cross validation. Assessed by five-fold validation, the random forest regression (RF) and radial basis function (RBF) models demonstrated better prediction performance in eight attempted machine learning models. The CLtot values predicted by the RF and RBF models were within two-fold of the observed values for 67.7 and 71.9% of cluster-split test set compounds, respectively, while the predictivity was worse in the time-split dataset. The predictivity of both models tended to be improved by incorporating in vitro parameters, unbound fraction in plasma (fu,p), and CLint. CLtot prediction utilizing in vitro CLint and the well-stirred model, correcting for the fraction unbound in blood, was substantially worse compared to machine learning approaches for the same cluster-split test set. The reason that CLtot is underestimated by IVIVE is not fully explained by considering the calculated microsomal unbound fraction (cfu,mic), extended clearance classification system (ECCS), and omitting high clearance compounds in excess of hepatic blood flow. The analysis suggests that in silico machine learning models may have the power to reduce reliance on or replace in vitro and in vivo studies for chemical structure optimization in early drug discovery.

Fast inspection of fruits using nuclear magnetic resonance spectroscopy

AbstractHigh‐resolution nuclear magnetic resonance (NMR) spectroscopy is an indispensable technique for obtaining chemical structure information. Its quantitative and noninvasive properties have led to its growing popularity as an analytical tool in many fields, including biology, chemistry, medicine, and food science. During transportation and storage, chemical reactions among the many nutrients lead to a loss of food quality. In these circumstances, portable NMR spectrometers can readily be used for food inspection and quality control. Because of the heterogeneous tissue distribution in food, a high‐resolution NMR method is required for detailed food inspection. Therefore, in this study, we demonstrated the feasibility of using an intermolecular double‐quantum coherence signal to obtain high‐resolution metabolic profiles of several fruits, including grape, cantaloupe, tomato, and watermelon. The resulting high‐resolution NMR spectra facilitate the identification of important metabolites, which can be used as biomarkers for food quality control. The method established here may be adapted for food inspection using portable NMR equipment.

Incorporating Clinical, Chemical and Biological Information for Predicting Small Molecule-microRNA Associations Based on Non-Negative Matrix Factorization.

Small molecule(SM) drugs can affect the expression of miRNAs, which plays crucial roles in many important biological processes. The chemical structure and clinical information of small molecule can simultaneously incorporate information such as anatomical distribution, therapeutic effects and structural characteristics. It is necessary to develop a novel model that incorporates small molecule chemical structure and clinical information to reveal the unknown small molecule-miRNA associations. In this study, we developed a new framework based on non-negative matrix factorization, called SMANMF, to discover the potential small molecules-miRNAs associations. First, the functional similarity of two miRNAs can be obtained by computing the overlap of the target gene sets in which the miRNAs interact together, and we integrated two types of small molecule similarities, including chemical similarity and clinical similarity. Then, we utilized a non-negative matrix factorization model to discover the unknown relationship between small molecules and miRNAs. The evaluation results indicate that our model can achieve superior prediction performance compared with previous approaches in 5-fold cross-validation. At the same time, the results of case studies also reveal that the SMANMF model has good predictive performance for predicting the potential association between small molecules and miRNAs.

Exploration and Evaluation of Machine Learning-Based Models for Predicting Enzymatic Reactions.

Unannotated gene sequences in databases are increasing due to sequencing advances. Therefore, computational methods to predict functions of unannotated genes are needed. Moreover, novel enzyme discovery for metabolic engineering applications further encourages annotation of sequences. Here, enzyme functions are predicted using two general approaches, each including several machine learning algorithms. First, Enzyme-models (E-models) predict Enzyme Commission (EC) numbers from amino acid sequence information. Second, Substrate-Enzyme models (SE-models) are built to predict substrates of enzymatic reactions together with EC numbers, and Substrate-Enzyme-Product models (SEP-models) are built to predict substrates, products, and EC numbers. While accuracy of E-models is not optimal, SE-models and SEP-models predict EC numbers and reactions with high accuracy using all tested machine learning-based methods. For example, a single Random Forests-based SEP-model predicts EC first digits with an Average AUC score of over 0.94. Various metrics indicate that the current strategy of combining sequence and chemical structure information is effective at improving enzyme reaction prediction.

Molecular description for magnesium iodide

Graph theory, which has become a powerful area of mathematics, owns much advancement in the field of mathematical chemistry. Recently, chemical graph theory has turned into a very popular area among researchers because of its wide-ranging applications in the field of mathematical chemistry. The manipulation and inspection of chemical structural information is made feasible using molecular descriptors. The molecular topological descriptors are the numerical invariants of a molecular graph and are valuable for predicting their bioactivity. An abundant variety of such indices are taken into consideration and used in pharmaceutical researchers, in theoretical chemistry, in drugs and in several other fields. A topological index actually relates a chemical structure by means of a numeric number. In this recent research work, we have considered the chemical graph of magnesium iodide. We computed degree based topological indices. Mainly, we addressed atom-bond connectivity index (ABC), geometric arithmetic index (GA), fourth atom-bond connectivity index (ABC4), The fifth geometric-arithmetic index (GA5), general Randic' index Ra(G) and First Zagreb index M1(G), Second Zagreb index M2(G) for magnesium iodide, MgI2. Furthermore, the results are analysed and we have provided general formulas for all these above mentioned families of graphs that are in fact very helpful in studying the underlying topologies.

Development of an in silico prediction system of human renal excretion and clearance from chemical structure information incorporating fraction unbound in plasma as a descriptor

Prediction of pharmacokinetic profiles of new chemical entities is essential in drug development to minimize the risks of potential withdrawals. The excretion of unchanged compounds by the kidney constitutes a major route in drug elimination and plays an important role in pharmacokinetics. Herein, we created in silico prediction models of the fraction of drug excreted unchanged in the urine (fe) and renal clearance (CLr), with datasets of 411 and 401 compounds using freely available software; notably, all models require chemical structure information alone. The binary classification model for fe demonstrated a balanced accuracy of 0.74. The two-step prediction system for CLr was generated using a combination of the classification model to predict excretion-type compounds and regression models to predict the CLr value for each excretion type. The accuracies of the regression models increased upon adding a descriptor, which was the observed and predicted fraction unbound in plasma (fu,p); 78.6% of the samples in the higher range of renal clearance fell within 2-fold error with predicted fu,p value. Our prediction system for renal excretion is freely available to the public and can be used as a practical tool for prioritization and optimization of compound synthesis in the early stage of drug discovery.

Amorphous Carbon Films In Situ Formed From PTFE Transfer Layer in Solid Lubricated Cryogenic Turbopump Bearings

In order to fully evaluate the service reliability of cryogenic turbopump bearings, a newly developed solid lubricated bearing was tested using a custom ground bearing test system cooled by water. Stains with different color intensity were occasionally exhibited on the surfaces of the bearing balls after ground tests. This may pose a risk for the reliable service of the related bearings in real flight missions. Then the chemical composition and structural information of the stain layers were systematically analyzed by x-ray photoelectron spectroscopy (XPS), Raman spectroscopy, and high-resolution transmission electron microscope (HRTEM). The results manifested that the stain layer had the characteristic of α-C film. The color intensity of stains was not caused by the structural changes, but was due to the difference of stain layers’ thickness. The α-C films in situ formed during bearings operation should result from the defluorination of polytetrafluoroethylene (PTFE) layers, which transferred from retainer to balls in the previous step. Correspondingly, FeF2 was produced on the surfaces of the tested bearing rings as a by-product. Although the tested bearings with α-C films formed on the surfaces of balls did not reach the state of final failure, the affected balls showed high surface roughness, which could be an indicator of an imminent bearing failure.

Online Characterization of Single Airborne Carbon Nanotube Particles Using Optical Trapping Raman Spectroscopy.

Carbon nanotubes (CNTs) have become recognized as a potential environmental and health hazard as their applications are broadening and manufacturing costs are reducing. Fundamental information of CNTs in air is of significant importance to our understanding of their environmental fate as well as to further applications. Extensive efforts have been made over decades on characterizing CNTs; however, a majority of the studies are of bulk or CNTs dispersed on substrates. In the present study, we characterize single CNT particles in air using optical trapping Raman spectroscopy (OT-RS). Different types of CNT particles, as well as glassy carbon spheres, were optically trapped in air. Their physical properties were viewed by microscopic bright field images and scattering images; their chemical properties and structural information can be inferred from characteristic Raman bands. The system can also spatially resolve the morphology and chemical distribution of optically trapped CNT particles in air. The OT-RS technique combines single-particle morphological and chemical information and offers an online method to characterize the physicochemical properties of single CNT particles at their native states in air.

Development of Simplified in Vitro P-Glycoprotein Substrate Assay and in Silico Prediction Models To Evaluate Transport Potential of P-Glycoprotein.

For efficient drug discovery and screening, it is necessary to simplify P-glycoprotein (P-gp) substrate assays and to provide in silico models that predict the transport potential of P-gp. In this study, we developed a simplified in vitro screening method to evaluate P-gp substrates by unidirectional membrane transport in P-gp-overexpressing cells. The unidirectional flux ratio positively correlated with parameters of the conventional bidirectional P-gp substrate assay ( R2 = 0.941) and in vivo Kp,brain ratio (mdr1a/1b KO/WT) in mice ( R2 = 0.800). Our in vitro P-gp substrate assay had high reproducibility and required approximately half the labor of the conventional method. We also constructed regression models to predict the value of P-gp-mediated flux and three-class classification models to predict P-gp substrate potential (low-, medium-, and high-potential) using 2397 data entries with the largest data set collected under the same experimental conditions. Most compounds in the test set fell within two- and three-fold errors in the random forest regression model (71.3 and 88.5%, respectively). Furthermore, the random forest three-class classification model showed a high balanced accuracy of 0.821 and precision of 0.761 for the low-potential classes in the test set. We concluded that the simplified in vitro P-gp substrate assay was suitable for compound screening in the early stages of drug discovery and that the in silico regression model and three-class classification model using only chemical structure information could identify the transport potential of compounds including P-gp-mediated flux ratios. Our proposed method is expected to be a practical tool to optimize effective central nervous system (CNS) drugs, to avoid CNS side effects, and to improve intestinal absorption.

XAFS Studies of Fe−SiO2 Fischer‐Tropsch Catalyst During Activation in CO, H2, and Synthesis Gas

AbstractHighly dispersed catalysts are widely used in industry and academia because of their large surface active sites. However, the predominant short‐range ordered structures in catalysts cannot be revealed by bulk characterization tools. The present study chooses a highly dispersed Fe−SiO2 FTS catalyst as a research object and shows how X‐ray absorption fine structure (XAFS) helps to analyze the chemical information (iron composition, reduction degree, and iron valence) and structural information (coordinating atom and coordination number) of catalyst. The activation process of catalysts are detailedly studied under different atmospheres (CO, H2, or 2H2/CO) by using both ex‐situ and in‐situ XAFS. CO activation behaves most effective during the whole process and finally produces the highest content of active iron species compared with H2 and 2H2/CO activation. The coordination number evolution of nearest Fe−O and Fe−C shells around Fe are put forward to understand the activation process. The present study demonstrates how XAFS elaborately reveals the phase transformations of the highly dispersed Fe−SiO2 FTS catalyst.

Modeling Antibacterial Activity with Machine Learning and Fusion of Chemical Structure Information with Microorganism Metabolic Networks.

Predicting the activity of new chemical compounds over pathogenic microorganisms with different metabolic reaction networks (MRN s) is an important goal due to the different susceptibility to antibiotics. The ChEMBL database contains >160 000 outcomes of preclinical assays of antimicrobial activity for 55 931 compounds with >365 parameters of activity (MIC, IC50, etc.) and >90 bacteria strains of >25 bacterial species. In addition, the Leong and Barabàsi data set includes >40 MRNs of microorganisms. However, there are no models able to predict antibacterial activity for multiple assays considering both drug and MRN structures at the same time. In this work, we combined perturbation theory, machine learning, and information fusion techniques to develop the first PTMLIF model. The best linear model found presented values of specificity = 90.31/90.40 and sensitivity = 88.14/88.07 in training/validation series. We carried out a comparison to nonlinear artificial neural network (ANN) techniques and previous models from the literature. Next, we illustrated the practical use of the model with an experimental case of study. We reported for the first time the isolation and characterization of terpenes from the plant Cissus incisa. The antibacterial activity of the terpenes was experimentally determined. The more active compounds were phytol and α-amyrin, with MIC = 100 μg/mL for Vancomycin-resistant Enterococcus faecium and Acinetobacter baumannii resistant to carbapenems. These compounds are already known from other sources. However, they have been isolated and evaluated for the first time here against several strains of multidrug-resistant bacteria including World Health Organization (WHO) priority pathogens. Last, we used the model to predict the activity of these compounds versus other microorganisms with different MRNs in order to find other potential targets.

Surrogate Fuels Formulation for FACE Gasoline Using the Nuclear Magnetic Resonance Spectroscopy

An efficient surrogate fuel formulation methodology, which directly uses the chemical structure information from nuclear magnetic resonance (NMR) spectroscopy analysis, has been proposed. Five functional groups, paraffinic CH2, paraffinic CH3, aromatic C-CH, olefinic CH-CH2, and cycloparaffin CH2, have been selected to show the basic molecular structure of the fuels for the advanced combustion engines (FACE) fuels. A palette that contains six candidate components, n-heptane, iso-octane, toluene, 2,5-dimethylhexane, methylcyclohexane, and 1-hexene, is chosen for different FACE fuels, based on the consideration that surrogate mixtures should provide the representative functional groups and comparable molecular sizes. The kinetic mechanisms of these six candidate components are chosen to assemble a detailed mechanism of each surrogate fuel for FACE gasoline. Whereafter, the accuracy of FACE A and F surrogate models was demonstrated by comparing the model predictions against experimental data in homogeneous ignition, jet stirred reactor oxidation, and premixed flame.

A mechanistic framework for integrating chemical structure and high-throughput screening results to improve toxicity predictions

Adverse Outcome Pathways (AOPs) establish a connection between a molecular initiating event (MIE) and an adverse outcome. Detailed understanding of the MIE provides the ideal data for determining chemical properties required to elicit the MIE. This study utilized high-throughput screening data from the ToxCast program, coupled with chemical structural information, to generate chemical clusters using three similarity methods pertaining to nine MIEs within an AOP network for hepatic steatosis. Three case studies demonstrate the utility of the mechanistic information held by the MIE for integrating biological and chemical data. Evaluation of the chemical clusters activating the glucocorticoid receptor identified activity differences in chemicals within a cluster. Comparison of the estrogen receptor results with previous work showed that bioactivity data and structural alerts can be combined to improve predictions in a customizable way where bioactivity data are limited. The aryl hydrocarbon receptor (AHR) highlighted that while structural data can be used to offset limited data for new screening efforts, not all ToxCast targets have sufficient data to define robust chemical clusters. In this context, an alternative to additional receptor assays is proposed where assays for proximal key events downstream of AHR activation could be used to enhance confidence in active calls. These case studies illustrate how the AOP framework can support an iterative process whereby in vitro toxicity testing and chemical structure can be combined to improve toxicity predictions. In vitro assays can inform the development of structural alerts linking chemical structure to toxicity. Consequently, structurally related chemical groups can facilitate identification of assays that would be informative for a specific MIE. Together, these activities form a virtuous cycle where the mechanistic basis for the in vitro results and the breadth of the structural alerts continually improve over time to better predict activity of chemicals for which limited toxicity data exist.

Core-Shell Architectured High Energy Cathode Nanocrystals to Stabilize the Electrode-Electrolyte Interface

Stabilization of high energy cathode-electrolyte interface is required to meet the criteria for the power source and stable cycling performance in Li-ion batteries. [1] The undesired reactions at the interface could be minimized by introducing redox-inactive ions on the surface of active materials without affecting original properties. [2] To demonstrate the roles of passivating layers more effectively, high energy nanocrystals were introduced due to its high surface area that can induce access of charges with electrolytes, accelerating chemical degradation at the interface. [3] A synthetic method based on colloidal chemistry was employed to grow a thin shell enriched in Al3+ in the same reaction environment where the particles were grown in dispersible form, ensuring effective coverage of all surfaces. The resultant core-shell nanocrystal consists of active lithium transition metal oxides as a core and enriched Al3+ shells as a passivating layers to suppress the unfavorable reaction on the surface. Careful post-synthetic annealing was used to produce the final heterostructures and tailor the specific chemistry of the shells. The modified nanocrystals improved capacity retentions in various conditions, such as elevated temperature and high applying potentials compared to the bare counterpart where a shell is not presented, strongly suggesting a role of protective layers. The chemical identity and specific structural information of the bulk and passivation layers were mainly characterized by a combination of solid state nuclear magnetic resonance (NMR), X-ray photoelectron spectroscopy (XPS), scanning transmission electron microscopy (STEM), and X-ray absorption spectroscopy (XAS). Reference Lee, K. T.; Jeong, S.; Cho, J. Acc. Chem. Res, 2013, 46, 1161-1170.Kim, C.; Phillips, P. J.; Xu, L. P.; Dong, A. G.; Buonsanti, R.; Klie, R. F.; Cabana, J. Chem. Mater, 2015, 27, 394-399.Bruce, P. G.; Scrosati, B.; Tarascon, J. M. Angew. Chem. Int. Ed, 2008, 47, 2930-2946.

(Invited) One Step Synthesis of Li-Alkyl Carbonates and Their Applications As Coatings on Li Anode

Lithium metal is being extensively studied as an anode to replace graphite due to its high capacity and the lowest negative electrochemical potential.1 Current research focus is to mitigate lithium dendrite growth and to increase coulomb efficiency during electrochemical cycling by using various surface coatings or electrolyte additives. Due to the complex reduction reactions of electrolyte components on lithium metal surface, it is highly desirable to develop coating materials with high electrochemical stability. In this work, we report the synthesis of several Li-alkyl carbonates via a simple one step reaction. These carbonates are known components of solid electrolyte interface (SEI) layers on lithium and graphite anode surfaces. The chemical composition and structural information of these synthetic Li-alkyl carbonates are identified by X-ray diffraction and Infrared Spectroscopy. In one implementation of the reaction, a single-component coating is formed on Li metal in a LiI/organic carbonates (dimethyl carbonate (DMC) and ethylene carbonate (EC)) electrolyte. LiI chemically reacts with dimethyl carbonate to form lithium methyl carbonate, which precipitates and forms a chemically homogeneous coating layer on Li surface. This coating layer is shown to enable dendrite free Li cycling in a symmetric Li||Li cell even at a current density of 3 mA cm-2. Adding EC to DMC modulates the formation of the lithium methyl carbonate, resulting in a stable coating layer which is essential for long term Li cycling stability. Fig.1 shows the morphology of deposited Li using LiI-EC/DMC electrolytes.2 Furthermore, the coating can enable dendrite free, high coulomb efficiency cycling after being transferred to common LiPF6/carbonates electrolytes which are compatible with metal oxide cathodes. We will show that LiI reacts with a broad range of organic carbonates and ethers to form lithium compounds relevant to high efficiency operations of battery anodes. Acknowledgements This work was supported by the Office of Vehicle Technologies of the U.S. Department of Energy through the Advanced Battery Materials Research Program (Battery 500 Consortium) under Contract DE-EE0007764. Reference W. Xu, J. L. Wang, F. Ding, X. L. Chen, E. Nasybutin, Y. H. Zhang and J. G. Zhang, Energ Environ Sci, 2014, 7, 513-537.H. D. Liu, H. Y. Zhou, B. S. Lee, X. Xing, M. Gonzalez and P. Liu, ACS Appl Mater Inter, 2017, 9, 30635-30642. Figure 1

IMPPAT: A curated database of Indian Medicinal Plants, Phytochemistry And Therapeutics

Phytochemicals of medicinal plants encompass a diverse chemical space for drug discovery. India is rich with a flora of indigenous medicinal plants that have been used for centuries in traditional Indian medicine to treat human maladies. A comprehensive online database on the phytochemistry of Indian medicinal plants will enable computational approaches towards natural product based drug discovery. In this direction, we present, IMPPAT, a manually curated database of 1742 Indian Medicinal Plants, 9596 Phytochemicals, And 1124 Therapeutic uses spanning 27074 plant-phytochemical associations and 11514 plant-therapeutic associations. Notably, the curation effort led to a non-redundant in silico library of 9596 phytochemicals with standard chemical identifiers and structure information. Using cheminformatic approaches, we have computed the physicochemical, ADMET (absorption, distribution, metabolism, excretion, toxicity) and drug-likeliness properties of the IMPPAT phytochemicals. We show that the stereochemical complexity and shape complexity of IMPPAT phytochemicals differ from libraries of commercial compounds or diversity-oriented synthesis compounds while being similar to other libraries of natural products. Within IMPPAT, we have filtered a subset of 960 potential druggable phytochemicals, of which majority have no significant similarity to existing FDA approved drugs, and thus, rendering them as good candidates for prospective drugs. IMPPAT database is openly accessible at: https://cb.imsc.res.in/imppat.

OPERA models for predicting physicochemical properties and environmental fate endpoints

The collection of chemical structure information and associated experimental data for quantitative structure–activity/property relationship (QSAR/QSPR) modeling is facilitated by an increasing number of public databases containing large amounts of useful data. However, the performance of QSAR models highly depends on the quality of the data and modeling methodology used. This study aims to develop robust QSAR/QSPR models for chemical properties of environmental interest that can be used for regulatory purposes. This study primarily uses data from the publicly available PHYSPROP database consisting of a set of 13 common physicochemical and environmental fate properties. These datasets have undergone extensive curation using an automated workflow to select only high-quality data, and the chemical structures were standardized prior to calculation of the molecular descriptors. The modeling procedure was developed based on the five Organization for Economic Cooperation and Development (OECD) principles for QSAR models. A weighted k-nearest neighbor approach was adopted using a minimum number of required descriptors calculated using PaDEL, an open-source software. The genetic algorithms selected only the most pertinent and mechanistically interpretable descriptors (2–15, with an average of 11 descriptors). The sizes of the modeled datasets varied from 150 chemicals for biodegradability half-life to 14,050 chemicals for logP, with an average of 3222 chemicals across all endpoints. The optimal models were built on randomly selected training sets (75%) and validated using fivefold cross-validation (CV) and test sets (25%). The CV Q2 of the models varied from 0.72 to 0.95, with an average of 0.86 and an R2 test value from 0.71 to 0.96, with an average of 0.82. Modeling and performance details are described in QSAR model reporting format and were validated by the European Commission’s Joint Research Center to be OECD compliant. All models are freely available as an open-source, command-line application called OPEn structure–activity/property Relationship App (OPERA). OPERA models were applied to more than 750,000 chemicals to produce freely available predicted data on the U.S. Environmental Protection Agency’s CompTox Chemistry Dashboard.

Expanding biological space coverage enhances the prediction of drug adverse effects in human using in vitro activity profiles

In vitro assay data have recently emerged as a potential alternative to traditional animal toxicity studies to aid in the prediction of adverse effects of chemicals on humans. Here we evaluate the data generated from a battery of quantitative high-throughput screening (qHTS) assays applied to a large and diverse collection of chemicals, including approved drugs, for their capacity in predicting human toxicity. Models were built with animal in vivo toxicity data, in vitro human cell-based assay data, as well as in combination with chemical structure and/or drug-target information to predict adverse effects observed for drugs in humans. Interestingly, we found that the models built with the human cell-based assay data performed close to those of the models based on animal in vivo toxicity data. Furthermore, expanding the biological space coverage of assays by including additional drug-target annotations was shown to significantly improve model performance. We identified a small set of targets, which, when added to the current suite of in vitro human cell-based assay data, result in models that greatly outperform those built with the existing animal toxicity data. Assays can be developed for this set of targets to screen compounds for construction of robust models for human toxicity prediction.