Development Of Quantitative Structure-activity Relationship Models Research Articles

Development of MLR and SVM Aided QSAR Models to Identify Common SAR of GABA Uptake Herbal Inhibitors used in the Treatment of Schizophrenia.

Background: Alterations in GABAnergic system are implicated in the pathophysiology of schizophrenia. Available antipsychotics that target GABA receptor form a desirable therapeutic strategy in the treatment regimen of schizophrenia, unfortunately, suffer serious setback due to their prolonged side effects. The present investigation focuses on developing QSAR models from the biological activity of herbal compounds and their derivatives that promise to be alternative candidates to GABA uptake inhibitors.Methods: Three sets of compounds were undertaken in the study to develop QSAR models. The first set consisted of nine compounds which included Magnolol, Honokiol and other GABA acting established compounds. The second set consisted of 16 derivatives of N-diarylalkenyl-piperidinecarboxylic acid. The third QSAR dataset was made up of thirty two compounds which were Magnolol and Honokiol derivatives. Multiple linear regressions (MLR) and support vector machine (SVM) supervised quantitative structure-activity relationship (QSAR) models were developed to predict the biological activity of these three sets. The purpose of taking three QSAR sets of diverse chemical structures but identical in their GABA targeting and pharmacological action was to identify common chemical structure features responsible for structure-activity relationship (SAR).Results: Linear and non-linear QSAR models confirmed that the three sets shared common structural descriptors derived from WHIM (Weighted Holistic Invariant Molecular descriptors), 3D-MoRSE and Eigenvalue classes.Conclusion: It was concluded that properties like electro negativity and polarizability play a crucial role in controlling the activity of herbal compounds against GABA receptor.

Predictive QSAR Models for the Toxicity of Disinfection Byproducts.

Several hundred disinfection byproducts (DBPs) in drinking water have been identified, and are known to have potentially adverse health effects. There are toxicological data gaps for most DBPs, and the predictive method may provide an effective way to address this. The development of an in-silico model of toxicology endpoints of DBPs is rarely studied. The main aim of the present study is to develop predictive quantitative structure–activity relationship (QSAR) models for the reactive toxicities of 50 DBPs in the five bioassays of X-Microtox, GSH+, GSH−, DNA+ and DNA−. All-subset regression was used to select the optimal descriptors, and multiple linear-regression models were built. The developed QSAR models for five endpoints satisfied the internal and external validation criteria: coefficient of determination (R2) > 0.7, explained variance in leave-one-out prediction (Q2LOO) and in leave-many-out prediction (Q2LMO) > 0.6, variance explained in external prediction (Q2F1, Q2F2, and Q2F3) > 0.7, and concordance correlation coefficient (CCC) > 0.85. The application domains and the meaning of the selective descriptors for the QSAR models were discussed. The obtained QSAR models can be used in predicting the toxicities of the 50 DBPs.

Adaptive neuro-fuzzy inference system-applied QSAR with bond dissociation energy for antioxidant activities of phenolic compounds.

The aim of this study was to develop quantitative structure-activity relationship (QSAR) models for predicting antioxidant activities of phenolic compounds. The bond dissociation energy of O-H bond (BDE) was calculated by semi-empirical quantum chemical methods. As a new parameter for QSAR models, sum of reciprocals of BDE of enol and phenol groups (X BDE ) was calculated. Significant correlations were observed between X BDE and antioxidant activities, and X BDE was introduced as a parameter for developing QSAR models. Linear regression-applied QSAR models and adaptive neuro-fuzzy inference system (ANFIS)-applied QSAR models were developed. QSAR models by both of linear regression and ANFIS achieved high prediction accuracies. Among the developed models, ANFIS-applied models achieved better prediction accuracies than linear regression-applied models. From these results, the proposed parameter of X BDE was confirmed as an appropriate variable for predicting and analysing antioxidant activities of phenolic compounds. Also, the ANFIS could be applied on QSAR models to improve prediction accuracy.

QSAR, docking, ADMET, and system pharmacology studies on tormentic acid derivatives for anticancer activity

To explore the anticancer compounds from tormentic acid derivatives, a quantitative structure–activity relationship (QSAR) model was developed by the multiple linear regression methods. The developed QSAR model yielded a high activity–descriptors relationship accuracy of 94% referred by regression coefficient (r2 = .94) and a high activity prediction accuracy of 91%. The QSAR study indicates that chemical descriptors, chiV5, T_T_Cl_7, T_2_T_4, SsCH3count, and Epsilon3 are significantly correlated with anticancer activity. This validated model was further been used for virtual screening and thus identification of new potential breast cancer inhibitors. Lipinski’s rule of five, ADMET risk and synthetic accessibility are used to filter false positive hits. Filtered compounds were then docked to identify the possible target binding pocket, to obtain a set of aligned ligand poses and to prioritize the predicted active compounds. The scrutinized compounds, as well as their metabolites, were predicted and analyzed for different pharmacokinetics parameters such as absorption, distribution, metabolism, excretion, and toxicity. Finally, the top-ranked compound NB-12 was evaluated by system pharmacology approach. Later studied the metabolic networks, disease biomarker networks, pathway maps, drug-target networks and generate significant gene networks. The strategy applied in this research work may act as a framework for rational design of potential anticancer drugs.

Predictive models for identifying the binding activity of structurally diverse chemicals to human pregnane X receptor.

Toxic chemicals entered into human body would undergo a series of metabolism, transport and excretion, and the key roles played in there processes were metabolizing enzymes, which was regulated by the pregnane X receptor (PXR). However, some chemicals in environment could activate or antagonize human pregnane X receptor, thereby leading to a disturbance of normal physiological systems. In this study, based on a larger number of 2724 structurally diverse chemicals, we developed qualitative classification models by the k-nearest neighbor method. Moreover, the logarithm of 20 and 50% effective concentrations (log EC 20 and log EC 50) was used to establish quantitative structure-activity relationship (QSAR) models. With the classification model, two descriptors were enough to establish acceptable models, with the sensitivity, specificity, and accuracy being larger than 0.7, highlighting a high classification performance of the models. With two QSAR models, the statistics parameters with the correlation coefficient (R 2) of 0.702-0.749 and the cross-validation and external validation coefficient (Q 2) of 0.643-0.712, this indicated that the models complied with the criteria proposed in previous studies, i.e., R 2>0.6, Q 2>0.5. The small root mean square error (RMSE) of 0.254-0.414 and the good consistency between observed and predicted values proved satisfactory goodness of fit, robustness, and predictive ability of the developed QSAR models. Additionally, the applicability domains were characterized by the Euclidean distance-based approach and Williams plot, and results indicated that the current models had a wide applicability domain, which especially included a few classes of environmental contaminant, those that were not included in the previous models.

Experimental Errors in QSAR Modeling Sets: What We Can Do and What We Cannot Do

Numerous chemical data sets have become available for quantitative structure–activity relationship (QSAR) modeling studies. However, the quality of different data sources may be different based on the nature of experimental protocols. Therefore, potential experimental errors in the modeling sets may lead to the development of poor QSAR models and further affect the predictions of new compounds. In this study, we explored the relationship between the ratio of questionable data in the modeling sets, which was obtained by simulating experimental errors, and the QSAR modeling performance. To this end, we used eight data sets (four continuous endpoints and four categorical endpoints) that have been extensively curated both in-house and by our collaborators to create over 1800 various QSAR models. Each data set was duplicated to create several new modeling sets with different ratios of simulated experimental errors (i.e., randomizing the activities of part of the compounds) in the modeling process. A fivefold cross-validation process was used to evaluate the modeling performance, which deteriorates when the ratio of experimental errors increases. All of the resulting models were also used to predict external sets of new compounds, which were excluded at the beginning of the modeling process. The modeling results showed that the compounds with relatively large prediction errors in cross-validation processes are likely to be those with simulated experimental errors. However, after removing a certain number of compounds with large prediction errors in the cross-validation process, the external predictions of new compounds did not show improvement. Our conclusion is that the QSAR predictions, especially consensus predictions, can identify compounds with potential experimental errors. But removing those compounds by the cross-validation procedure is not a reasonable means to improve model predictivity due to overfitting.

Impact of geometry optimization methods on QSAR modelling: A case study for predicting human serum albumin binding affinity

Quantitative structure–activity relationship (QSAR) modelling is a major tool employed in the prediction of various endpoints. However, current QSAR literature is missing a full understanding of the impact of quantum chemical calculation methods on the estimation of molecular descriptors and model performance. Here, we provide a comprehensive analysis of the quantitative effects of different geometry optimization methods (semi-empirical, ab initio Hartee-Fock and density functional theory) on the molecular descriptors. Using experimental binding affinity to human serum albumin (HSA) data, we comparatively investigated the influence of employing descriptors derived from three calculation methods on the QSAR models. We propose a 4-descriptor QSAR model in line with the OECD validation principles for the prediction of drug binding affinity to HSA (log KHSA) as a potential tool for drug development. We also confirm the prediction capability of the proposed model on a heterogeneous external set of chemicals. Furthermore, we recommend an activity-independent rational approach for the selection of geometry optimization method for an improved QSAR model development.

A new kernel function of support vector regression combined with probability distribution and its application in chemometrics and the QSAR modeling

Quantitative structure–activity relationship (QSAR) models are extensively used to identify new chemicals affecting human health and speed up the drug discovery process. The development of accurate QSAR models can lead to a reduced number of experiments conducted on rats and mice to analyze new compounds. In a typical QSAR model, only the relationship among variables is considered, and the probability distribution of the samples is disregarded. Thus, a new kernel function of support vector regression (SVR) that integrates probability distribution is proposed. The proposed kernel function, called SVR-pk, satisfies kernel function theory, and the mean and variance of the sample are used to reflect the main distribution information. To verify the performance of the new kernel function, simulation example, two sets of data from UCI (University of California, Irvine) and two experiments about the compounds toxicity in rodents data obtained from the Carcinogenic Potency Database are employed. Results show that compared with other SVR models utilizing kernel functions, SVR-pk exhibits better performance and is more suitable for QSAR model.

A QSAR STUDY OF SUBSTITUTED PYRAZOLINE DERIVATIVES AS POTENTIAL ANTI-TUBERCULOSIS AGENTS

A quantitative structure–activity relationship (QSAR) of a series of substituted pyrazoline derivatives, in regard to their anti-tuberculosis activity, has been studied using the partial least square (PLS) analysis method. QSAR model development of 64 pyrazoline derivatives was carried out to predict anti-tubercular activity. Partial least square analysis was applied to derive QSAR models, which were further evaluated for statistical significance and predictive power by internal and external validation. The best QSAR model with good external and internal predictivity for the training and test set has shown cross validation (q2) and external validation (pred_r2) values of 0.7426 and 0.7903, respectively. Two-dimensional QSAR analyses of such pyrazoline derivatives provide important structural insights for designing potent antituberculosis drugs.

QSAR models for prediction of chromatographic behavior of homologous Fab variants.

While quantitative structure activity relationship (QSAR) models have been employed successfully for the prediction of small model protein chromatographic behavior, there have been few reports to date on the use of this methodology for larger, more complex proteins. Recently our group generated focused libraries of antibody Fab fragment variants with different combinations of surface hydrophobicities and electrostatic potentials, and demonstrated that the unique selectivities of multimodal resins can be exploited to separate these Fab variants. In this work, results from linear salt gradient experiments with these Fabs were employed to develop QSAR models for six chromatographic systems, including multimodal (Capto MMC, Nuvia cPrime, and two novel ligand prototypes), hydrophobic interaction chromatography (HIC; Capto Phenyl), and cation exchange (CEX; CM Sepharose FF) resins. The models utilized newly developed "local descriptors" to quantify changes around point mutations in the Fab libraries as well as novel cluster descriptors recently introduced by our group. Subsequent rounds of feature selection and linearized machine learning algorithms were used to generate robust, well-validated models with high training set correlations (R2 > 0.70) that were well suited for predicting elution salt concentrations in the various systems. The developed models then were used to predict the retention of a deamidated Fab and isotype variants, with varying success. The results represent the first successful utilization of QSAR for the prediction of chromatographic behavior of complex proteins such as Fab fragments in multimodal chromatographic systems. The framework presented here can be employed to facilitate process development for the purification of biological products from product-related impurities by in silico screening of resin alternatives. Biotechnol. Bioeng. 2017;114: 1231-1240. © 2016 Wiley Periodicals, Inc.

Development of a QSAR model for predicting aqueous reaction rate constants of organic chemicals with hydroxyl radicals.

Reaction with hydroxyl radicals (˙OH) is an important removal pathway for organic pollutants in the aquatic environment. The aqueous reaction rate constant (kOH) is therefore an important parameter for fate assessment of aquatic pollutants. Since experimental determination fails to meet the requirement of being able to efficiently handle numerous organic chemicals at limited cost and within a relatively short period of time, in silico methods such as quantitative structure-activity relationship (QSAR) models are needed to predict kOH. In this study, a QSAR model with a larger and wider applicability domain as compared with existing models was developed. Following the guidelines for the development and validation of QSAR models proposed by the Organization for Economic Co-operation and Development (OECD), the model shows satisfactory performance. The applicability domain of the model has been extended and contained chemicals that have rarely been covered in most previous studies. The chemicals covered in the current model contain functional groups including [double bond splayed left]C[double bond, length as m-dash]C[double bond splayed right], -C[triple bond, length as m-dash]C-, -C6H5, -OH, -CHO, -O-, [double bond splayed left]C[double bond, length as m-dash]O, -C[double bond, length as m-dash]O(O)-, -COOH, -C[triple bond, length as m-dash]N, [double bond splayed left]N-, -NH2, -NH-C(O)-, -NO2, -N[double bond, length as m-dash]C-N[double bond splayed right], [double bond splayed left]N-N[double bond splayed right], -N[double bond, length as m-dash]N-, -S-, -S-S-, -SH, -SO3, -SO4, -PO4, and -X (F, Cl, Br, and I).

Structure-based QSAR, molecule design and bioassays of protease-activated receptor 1 inhibitors

Quantitative structure–activity relationship (QSAR) studies were performed on a series of protease-activated receptor 1 (PAR1) inhibitors to identify the key structural features responsible for their biological activity. Induced-fit docking (IFD) was used to explore the active mechanisms of all PAR1 inhibitors at the active pocket of PAR1, and the best plausible conformation was determined by IFD for further QSAR studies. Based on the best plausible conformation, structure-based descriptors and ligand descriptors incorporating the ligand–receptor interaction were calculated. The random forest method was used to select important descriptors and build the 2D-QSAR model. The results of the 2D-QSAR model gave a squared correlation coefficient (R2) of 0.937, a prediction squared correlation coefficient (R2pred) of 0.845 and a mean square error (MSE) of 0.056. Furthermore, a 3D-QSAR model was developed via topomer comparative molecular field analysis (Topomer CoMFA), resulting in an R2 of 0.938, a cross-validated Q2 of 0.503 and a R2pred of 0.758. Based on the developed QSAR model, Topomer search was used for virtual screening of the R2 fragment in lead-like inhibitors from the National Cancer Institute (NCI) database, which contains 260,000 molecules. Eighty-two compounds were designed with different R2 fragments, and four of these compounds were selected for further biological testing. All four compounds showed inhibitory potency against PAR1.

The “double cross-validation” software tool for MLR QSAR model development

Quantitative structure–activity relationship (QSAR) modeling is a widely used computational technique applied in various fields including rational drug design, toxicity and property prediction of chemicals and pharmaceuticals, environmental risk assessment and fate modeling. External validation is generally considered as the gold standard in evaluating the model predictivity performance, at least to a group of QSAR practitioners. External validation is commonly performed by employing the hold-out method, where the original dataset is divided into training and test sets; the training set is employed for model building and model selection, while the test set is solely used in model assessment. However, since the composition of the training set remains the same in this method, it is not certain that the resultant model is optimal as there may be a bias in descriptor selection. This problem is more evident for the multiple linear regression (MLR) models than more robust and generalized partial least squares (PLS) and principal component regression (PCR) models. Thus, employing double cross-validation technique could be a better choice, in which the training set is further divided into ‘n’ calibration and validation sets resulting in diverse compositions. In the present work, we have developed an open access “Double Cross-Validation (DCV)” software tool which can be used to perform multiple linear regression (MLR) model development by employing the double cross-validation technique. Two variable selection methods, namely, stepwise MLR (S-MLR) and genetic algorithm MLR (GA-MLR) are incorporated in this tool and optionally, this tool also performs a data-pretreatment prior to the application of double cross-validation. Also, we have performed a study using the “Double Cross-Validation” tool on three different datasets in order to find out which technique among the hold-out and double cross-validation performs better in the selection of an optimal model in terms of model predictive performance checked on the test set. The performance of the tool in generating predictive PLS models is also compared. The “Double Cross-Validation” (version 2.0) tool is freely available to download from the sites http://teqip.jdvu.ac.in/QSAR_Tools/ and http://dtclab.webs.com/software-tools.

Development of TLSER model and QSAR model for predicting partition coefficients of hydrophobic organic chemicals between low density polyethylene film and water

Partition coefficients are vital parameters for measuring accurately the chemicals concentrations by passive sampling devices. Given the wide use of low density polyethylene (LDPE) film in passive sampling, we developed a theoretical linear solvation energy relationship (TLSER) model and a quantitative structure-activity relationship (QSAR) model for the prediction of the partition coefficient of chemicals between LDPE and water (Kpew). For chemicals with the octanol–water partition coefficient (log Kow) <8, a TLSER model with Vx (McGowan volume) and qA− (the most negative charge on O, N, S, X atoms) as descriptors was developed, but the model had relatively low determination coefficient (R2) and cross-validated coefficient (Q2). In order to further explore the theoretical mechanisms involved in the partition process, a QSAR model with four descriptors (MLOGP (Moriguchi octanol-water partition coeff.), P_VSA_s_3 (P_VSA-like on I-state, bin 3), Hy (hydrophilic factor) and NssO (number of atoms of type ssO)) was established, and statistical analysis indicated that the model had satisfactory goodness-of-fit, robustness and predictive ability. For chemicals with log KOW>8, a TLSER model with Vx and a QSAR model with MLOGP as descriptor were developed. This is the first paper to explore the models for highly hydrophobic chemicals. The applicability domain of the models, characterized by the Euclidean distance-based method and Williams plot, covered a large number of structurally diverse chemicals, which included nearly all the common hydrophobic organic compounds. Additionally, through mechanism interpretation, we explored the structural features those governing the partition behavior of chemicals between LDPE and water.

QSAR model development for studying carbonic anhydrase inhibitors as anticonvulsant agents

The objective of this work was to develop quantitative structure activity relationship (QSAR) models from sulfonamide derivatives against anticonvulsant activity. For developing the model, multiple linear regression and artificial neural network (ANN) have been employed as effective and efficient methods and these models have been validated with statistical analysis such as fraction of variance, cross-validation test, quality factor, Fischer’s test, and internal validation test (Y-randomization test), where applicable. A regression-based QSAR model (linear model) has been developed with cross-validation test q 2 = 0.8324 and fraction of variance r 2 = 0.8327, all the statistical tests have validated this model. An ANN (nonlinear model)-based model has also been developed with fraction of variance r 2 = 0.8710 and cross validation test q 2 = 0.7032. So, with the help of the developed models we can predict the logKi values of novel designed molecules and alter their structural properties accordingly before synthesizing them.

Discovery of Novel HIV-1 Integrase Inhibitors Using QSAR-Based Virtual Screening of the NCI Open Database.

Quantitative structure-activity relationship (QSAR) models can be used as a predictive tool for virtual screening of chemical libraries to identify novel drug candidates. The aims of this paper were to report the results of a study performed for descriptor selection, QSAR model development, and virtual screening for identifying novel HIV-1 integrase inhibitor drug candidates. First, three evolutionary algorithms were compared for descriptor selection: differential evolution-binary particle swarm optimization (DE-BPSO), binary particle swarm optimization, and genetic algorithms. Next, three QSAR models were developed from an ensemble of multiple linear regression, partial least squares, and extremely randomized trees models. A comparison of the performances of three evolutionary algorithms showed that DE-BPSO has a significant improvement over the other two algorithms. QSAR models developed in this study were used in consensus as a predictive tool for virtual screening of the NCI Open Database containing 265,242 compounds to identify potential novel HIV-1 integrase inhibitors. Six compounds were predicted to be highly active (plC50 > 6) by each of the three models. The use of a hybrid evolutionary algorithm (DE-BPSO) for descriptor selection and QSAR model development in drug design is a novel approach. Consensus modeling may provide better predictivity by taking into account a broader range of chemical properties within the data set conducive for inhibition that may be missed by an individual model. The six compounds identified provide novel drug candidate leads in the design of next generation HIV- 1 integrase inhibitors targeting drug resistant mutant viruses.

Imidazolium Ionic Liquids as Potential Anti-Candida Inhibitors: QSAR Modeling and Experimental Studies.

Quantitative structure-activity relationships (QSAR) of imidazolium ionic liquids (ILs) as inhibitors of C. albicans collection strains (IOA-109, KCTC 1940, ATCC 10231) have been studied. Predictive QSAR models were built using different descriptor sets for a set of 88 ionic liquids with known minimum inhibitory concentrations (MIC) against C. albicans. We applied the state-of-the-art QSAR methodologies such as WEKA Random Forest (RF) as a binary classifier, Associative Neural Networks (ASNN) and k-Nearest Neighbors (k-NN) to build continuum non-linear regression models. The obtained models were validated using a 5-fold cross-validation approach and resulted in the prediction accuracies of 80% ± 5.0 for the classification models and q2 = 0.73-0.87 for the non-linear regression models. Biological testing of newly synthesized 1,3-dialkylimidazolium ionic liquids with predicted activity was performed by disco-diffusion method against C. albicans ATCC 10231 M885 strain and clinical isolates C. albicans, C. krusei and C. glabrata strains. The high percentage of coincidence between the QSAR predictions and the experimental results confirmed the high predictive power of the developed QSAR models within the applicability domain of new imidazolium ionic liquids.

Quantitative structure–activity relationship model for prediction study of corrosion inhibition efficiency using two‐stage sparse multiple linear regression

AbstractA new quantitative structure–activity relationship (QSAR) of the inhibition of mild steel corrosion in 1 M hydrochloric acid using furan derivatives was developed by proposing two‐stage sparse multiple linear regression. The sparse multiple linear regression using ridge penalty and sparse multiple linear regression using elastic net (SMLRE) were used to develop the QSAR model. The results show that the SMLRE‐based model possesses high predictive power compared with sparse multiple linear regression using ridge penalty‐based model according to the mean‐squared errors for both training and test datasets, leave‐one‐out internal validation (Q2int = 0.98), and external validation (Q2ext = 0.95). In addition, the results of applicability domain assessment using the leverage approach reveal a reliable and robust SMLRE‐based model. In conclusion, the developed QSAR model using SMLRE can be efficiently used in the studies of corrosion inhibition efficiency. Copyright © 2016 John Wiley &amp; Sons, Ltd.

Development of classification model and QSAR model for predicting binding affinity of endocrine disrupting chemicals to human sex hormone-binding globulin

Disturbing the transport process is a crucial pathway for endocrine disrupting chemicals (EDCs) to disrupt endocrine function. However, this mechanism has not gotten enough attention, compared with that of hormone receptors and synthetase up to now, especially for the sex hormone transport process. In this study, we selected sex hormone-binding globulin (SHBG) and EDCs as a model system and the relative competing potency of a chemical with testosterone binding to SHBG (log RBA) as the endpoints, to develop classification models and quantitative structure-activity relationship (QSAR) models. With the classification model, a satisfactory model with nR09, nR10 and RDF155v as the most relevant variables was screened. Statistic results indicated that the model had the sensitivity, specificity, accuracy of 86.4%, 80.0%, 84.4% and 85.7%, 87.5%, 86.2% for the training set and validation set, respectively, highlighting a high classification performance of the model. With the QSAR model, a satisfactory model with statistical parameters, specifically, an adjusted determination coefficient (Radj2) of 0.810, a root mean square error (RMSE) of 0.616, a leave-one-out cross-validation squared correlation coefficient (QLOO2) of 0.777, a bootstrap method (QBOOT2) of 0.756, an external validation coefficient (Qext2) of 0.544 and a RMSEext of 0.859, were obtained, which implied satisfactory goodness of fit, robustness and predictive ability. The applicability domain of the current model covers a large number of structurally diverse chemicals, especially a few classes of nonsteroidal compounds.

Predictive Modeling of Estrogen Receptor Binding Agents Using Advanced Cheminformatics Tools and Massive Public Data

Estrogen receptors (ERα) are a critical target for drug design as well as a potential source of toxicity when activated unintentionally. Thus, evaluating potential ERα binding agents is critical in both drug discovery and chemical toxicity areas. Using computational tools, e.g., Quantitative Structure-Activity Relationship (QSAR) models, can predict potential ERα binding agents before chemical synthesis. The purpose of this project was to develop enhanced predictive models of ERα binding agents by utilizing advanced cheminformatics tools that can integrate publicly available bioassay data. The initial ERα binding agent data set, consisting of 446 binders and 8307 non-binders, was obtained from the Tox21 Challenge project organized by the NIH Chemical Genomics Center (NCGC). After removing the duplicates and inorganic compounds, this data set was used to create a training set (259 binders and 259 non-binders). This training set was used to develop QSAR models using chemical descriptors. The resulting models were then used to predict the binding activity of 264 external compounds, which were available to us after the models were developed. The cross-validation results of training set [Correct Classification Rate (CCR) = 0.72] were much higher than the external predictivity of the unknown compounds (CCR = 0.59). To improve the conventional QSAR models, all compounds in the training set were used to search PubChem and generate a profile of their biological responses across thousands of bioassays. The most important bioassays were prioritized to generate a similarity index that was used to calculate the biosimilarity score between each two compounds. The nearest neighbors for each compound within the set were then identified and its ERα binding potential was predicted by its nearest neighbors in the training set. The hybrid model performance (CCR = 0.94 for cross validation; CCR = 0.68 for external prediction) showed significant improvement over the original QSAR models, particularly for the activity cliffs that induce prediction errors. The results of this study indicate that the response profile of chemicals from public data provides useful information for modeling and evaluation purposes. The public big data resources should be considered along with chemical structure information when predicting new compounds, such as unknown ERα binding agents.