Applicability Domain Analysis Research Articles

Machine Learning Quantitative Structure-Property Relationships as a Function of Ionic Liquid Cations for the Gas-Ionic Liquid Partition Coefficient of Hydrocarbons.

Ionic liquids (ILs) are known for their unique characteristics as solvents and electrolytes. Therefore, new ILs are being developed and adapted as innovative chemical environments for different applications in which their properties need to be understood on a molecular level. Computational data-driven methods provide means for understanding of properties at molecular level, and quantitative structure–property relationships (QSPRs) provide the framework for this. This framework is commonly used to study the properties of molecules in ILs as an environment. The opposite situation where the property is considered as a function of the ionic liquid does not exist. The aim of the present study was to supplement this perspective with new knowledge and to develop QSPRs that would allow the understanding of molecular interactions in ionic liquids based on the structure of the cationic moiety. A wide range of applications in electrochemistry, separation and extraction chemistry depends on the partitioning of solutes between the ionic liquid and the surrounding environment that is characterized by the gas-ionic liquid partition coefficient. To model this property as a function of the structure of a cationic counterpart, a series of ionic liquids was selected with a common bis-(trifluoromethylsulfonyl)-imide anion, [Tf2N]−, for benzene, hexane and cyclohexane. MLR, SVR and GPR machine learning approaches were used to derive data-driven models and their performance was compared. The cross-validation coefficients of determination in the range 0.71–0.93 along with other performance statistics indicated a strong accuracy of models for all data series and machine learning methods. The analysis and interpretation of descriptors revealed that generally higher lipophilicity and dispersion interaction capability, and lower polarity in the cations induces a higher partition coefficient for benzene, hexane, cyclohexane and hydrocarbons in general. The applicability domain analysis of models concluded that there were no highly influential outliers and the models are applicable to a wide selection of cation families with variable size, polarity and aliphatic or aromatic nature.

In Silico Searching for Alternative Lead Compounds to Treat Type 2 Diabetes through a QSAR and Molecular Dynamics Study.

Free fatty acid receptor 1 (FFA1) stimulates insulin secretion in pancreatic β-cells. An advantage of therapies that target FFA1 is their reduced risk of hypoglycemia relative to common type 2 diabetes treatments. In this work, quantitative structure–activity relationship (QSAR) approach was used to construct models to identify possible FFA1 agonists by applying four different machine-learning algorithms. The best model (M2) meets the Tropsha’s test requirements and has the statistics parameters R2 = 0.843, Q2CV = 0.785, and Q2ext = 0.855. Also, coverage of 100% of the test set based on the applicability domain analysis was obtained. Furthermore, a deep analysis based on the ADME predictions, molecular docking, and molecular dynamics simulations was performed. The lipophilicity and the residue interactions were used as relevant criteria for selecting a candidate from the screening of the DiaNat and DrugBank databases. Finally, the FDA-approved drugs bilastine, bromfenac, and fenofibric acid are suggested as potential and lead FFA1 agonists.

Computer-aided estimation of kinetic rate constant for degradation of volatile organic compounds by hydroxyl radical: An improved model using quantum chemical and norm descriptors

The kinetic rate constant of volatile organic compounds (VOCs) degradation represents an important parameter, which is valuable for evaluating the removal efficiency and ecological risk of pollutants. In this study, the multiple-linear-regression method using quantum chemical and norm descriptors is utilized to develop a room-temperature quantitative structure–property relationships (QSPR) model for kinetic rate constant estimation. The correlation coefficient (R2) and root-mean-square error (RMSE) are 0.8918 and 0.4086 for the training set, as well as 0.9096 and 0.3901 for the test set, respectively, which suggests the as-developed model has good stability and predictability. Applicability domain analysis demonstrates that the model is reliable and generalizable for assessing the −logkOH of VOCs covering a wide variety of molecular structures. In addition, an external prediction is made to assess the degradation rate constants of nine hydrofluoroethers, which implies the predictability of the model. It is worth noting that the quantum mechanical parameters, i.e., natural population analysis and orbital energy for atoms are introduced to norm descriptors, which expands the number/type of norm descriptors and greatly improves the accuracy of the model. Such combinational quantum chemical and norm descriptors are expected to be used for building accurate and robust models for other chemical properties prediction.

Identification of concealed structural alerts using QSTR modeling for Pseudokirchneriella subcapitata.

In the present work, QSTR modeling was conducted for microalga Pseudokirchneriella subcapitata using a data set of 271 molecules belonging to different types of chemical classes for the prediction of EC50 for 72 hr based assays. The balanced QSTR model encompasses seven easily interpretable molecular descriptors and possesses statistical robustness with high predictive ability. This Genetic Algorithm Multi-linear regression (GA-MLR) model was subjected to internal validation, Y-randomization test, applicability domain analysis, and external validation as per the recommended OECD guidelines. The newly developed model fulfilled the threshold values for more than 20 recommended validation parameters including R2=0.72, Q2LOO=0.70, etc. The developed QSTR model was successful in identifying the type of hybridization or specific type of atoms of previously reported and newer structural alerts. Thus, the model could be useful for data gap filling and expanding mechanistic interpretation of toxicity for different chemicals.

Predicting melting point of ionic liquids using QSPR approach: Literature review and new models

Quantitative structure–property relationships (QSPRs) for predicting melting point temperature (Tm) of ionic liquids (ILs) are reviewed and the new models are proposed by using the experimental data extracted from the literature for 953 salts. The models include both regression of Tm data and classification of the ILs with respect to their state of matter (liquid/solid) at T=300 K. A variety of machine learning algorithms is applied, including: partial least squares regression, stepwise multiple linear regression, and a number of common classifiers (k-nearest neighbors, naive Bayes, linear discriminant analysis, support vector machines). An effect of the molecular descriptors set as well as the computational level used for the ions’ geometry optimization is analyzed and followed in the final model selection protocol, which comprises all the standard steps of good practice of QSRP modeling, e.g. cross-validation, external validation, and the applicability domain analysis. Furthermore, as a key novelty, the robustness of the models is checked for different combining rules, defined as the averaging functions for obtaining the descriptors of ILs from those given for individual ions. The finally selected and recommended models are discussed in detail in terms of various statistics, as well as addressed to other methods reported in the literature. An effect of the chemical family of both cation and anion on the modeling performance is highlighted. Additionally, the predictions of both Tm and state of matter of more than 35,000 virtual cation–anion combinations are given in order to present the range of potential applications of the new methods in computer-aided molecular design of new ILs displaying demanded phase behavior.

Quantitative structure-activity relationship study on the MMP-13 inhibitory activity of fused pyrimidine derivatives possessing a 1,2,4-Triazol-3-yl group as a ZBG

QSAR study has been carried out on the MMP-13 inhibitory activity of fused pyrimidine derivatives possessing a1,2,4-triazol-3-yl group as a ZBG in 0D- to 2D-Dragon descriptors. The derived QSAR models have revealed that the number of Sulfur atoms (descriptor nS), Balaban mean square distance index (descriptor MSD), molecular electrotopological variation (descriptor DELS), structural information content index of neighborhood symmetry of 2nd and 3rd order (descriptors SIC2 and SIC3), average valence connectivity index chi-4 (descriptor X4Av) in addition to 1st order Galvez topological charge index (descriptor JGI1) and global topological charge index (descriptor JGT) played a pivotal role in rationalization of MMP-13 inhibition activity of titled compounds. Atomic properties such as mass and volume in terms of atomic properties weighted descriptors MATS5m and MATS3v, and certain atom centred fragments such as CH2RX (descriptor C-006), X--CX--X (descriptor C-044), H attached to heteroatom (descriptor H-050) and H attached to C0(sp3) with 1X attached to next C (descriptor H-052) are also predominant to explain MMP-13 inhibition actions of fused pyrimidines. PLS analysis has also corroborated the dominance of CP-MLR identified descriptors. Applicability domain analysis revealed that the suggested model matches the high-quality parameters with good fitting power and the capability of assessing external data and all of the compounds was within the applicability domain of the proposed model and were evaluated correctly.

Alignment-Free Antimicrobial Peptide Predictors: Improving Performance by a Thorough Analysis of the Largest Available Data Set.

In the last two decades, a large number of machine-learning-based predictors for the activities of antimicrobial peptides (AMPs) have been proposed. These predictors differ from one another in the learning method and in the training and testing data sets used. Unfortunately, the training data sets present several drawbacks, such as a low representativeness regarding the experimentally validated AMP space, and duplicated peptide sequences between negative and positive data sets. These limitations give a low confidence to most of the approaches to be used in prospective studies. To address these weaknesses, we propose novel modeling and assessing data sets from the largest experimentally validated nonredundant peptide data set reported to date. From these novel data sets, alignment-free quantitative sequence-activity models (AF-QSAMs) based on Random Forest are created to identify general AMPs and their antibacterial, antifungal, antiparasitic, and antiviral functional types. An applicability domain analysis is carried out to determine the reliability of the predictions obtained, which, to the best of our knowledge, is performed for the first time for AMP recognition. A benchmarking is undertaken between the models proposed and several models from the literature that are freely available in 13 programs (ClassAMP, iAMP-2L, ADAM, MLAMP, AMPScanner v2.0, AntiFP, AMPfun, PEPred-suite, AxPEP, CAMPR3, iAMPpred, APIN, and Meta-iAVP). The models proposed are those with the best performance in all of the endpoints modeled, while most of the methods from the literature have weak-to-random predictive agreements. The models proposed are also assessed through Y-scrambling and repeated k-fold cross-validation tests, demonstrating that the outcomes obtained by them are not given by chance. Three chemometric analyses also confirmed the relevance of the peptides descriptors used in the modeling. Therefore, it can be concluded that the models built by fixing the drawbacks existing in the literature contribute to identifying antibacterial, antifungal, antiparasitic, and antiviral peptides with high effectivity and reliability. Models are freely available via the AMPDiscover tool at https://biocom-ampdiscover.cicese.mx/.

Development of blood brain barrier permeation prediction models for organic and inorganic biocidal active substances

Biocidal products are broadly used in homes and industries. However, the safety of biocidal active substances (BASs) is not yet fully understood. In particular, the neurotoxic action of BASs needs to be studied as diverse epidemiological studies have reported associations between exposure to BASs and neural diseases. In this study, we developed in silico models to predict the blood-brain barrier (BBB) permeation of organic and inorganic BASs. Due to a lack of BBB data for BASs, the chemical space of BASs and BBB dataset were compared in order to select BBB data that were structurally similar to BASs. In silico models to predict log-scaled BBB penetration were developed using support vector regression for organic BASs and multiple linear regression for inorganic BASs. The model for organic BASs was developed with 231 compounds (training set: 153 and test set: 78) and achieved good prediction accuracy on an external test set (R2 = 0.64), and the model outperformed the model for pharmaceuticals. The model for inorganic BASs was developed with 11 compounds (R2 = 0.51). Applicability domain (AD) analysis of the models clarified molecular structures reliably predicted by the models. Therefore, the models developed in this study can be used for predicting BBB permeable BASs in human. These models were developed according to the Quantitative Structure-Activity Relationship validation principles proposed by the Organization for Economic Cooperation and Development.

The chemical reactivity and QSPR of organic compounds applied to dye-sensitized solar cells using DFT

The structural and electronic properties were calculated for seventy organic compounds used as dye sensitizers in solar cells, applying the B3LYP exchange-correlation energy functional with the 6-311G∗∗ basis set. Moreover, the present study proposes two new quantitative structure-property relationship (QSPR) models that enable the prediction of the power conversion efficiency (PCE) and maximum absorption wavelength (λmax) of these systems, the two QSPR models were validated using the coefficient of determination (R2) of 0.62 for both models with the leave-one-out cross-validation correlation coefficient (Q2LOO) of 0.55 and 0.57, respectively. Furthermore, applicability domain analysis was conducted in order to identify the related compounds via the extrapolation of the model.

Molecular image-convolutional neural network (CNN) assisted QSAR models for predicting contaminant reactivity toward OH radicals: Transfer learning, data augmentation and model interpretation

In this study, we used molecular images as a representation for organic compounds and combined them with a convolutional neural network (CNN) to develop quantitative structure-activity relationships (QSARs) for predicting compound rate constants toward OH radicals. We applied transfer learning and data augmentation to train molecular image-CNN models and the Gradient-weighted Class Activation Mapping (Grad-CAM) method to interpret them. Results showed that data augmentation and transfer learning can effectively enhance the robustness and predictive performance of the models, with the root-mean-square-error (RMSE) values on the test dataset (RMSEtest) decreasing from (0.395–0.45) to (0.284–0.339) after applying data augmentation, and the RMSE on the training dataset (RMSEtrain) decreasing from (0.452–0.592) to (0.123–0.151) after applying transfer learning. The obtained molecular image-CNN models showed comparative predictive performance (RMSEtest 0.284–0.339) with the molecular fingerprint-based models (RMSEtest 0.30–0.35). Grad-CAM interpretation showed that the molecular image-CNN models correctly chose the molecular features in the images and identified key functional groups that influenced the reactivity. The applicability domain analysis showed that the molecular image-CNN models have a broader applicability domain than molecular fingerprints-based models and the reactivity of any new compounds with a maximum similarity of over 0.85 to the compounds in the training dataset can be reliably predicted. This study demonstrated that molecular image-CNN is a new tool to develop QSARs for environmental applications and can be used to build trustful models that make meaningful predictions.

A norm index-based QSPR model to predict the standard absolute entropy of organic compounds in three phase states

The entropy of organics is a fundamental and crucial thermodynamic quantity. In this work, a unified quantitative structure-property relationship (QSPR) model was proposed to predict the standard absolute entropy for 15 kinds of organic compounds in three phase states based on the norm index concept. Three phase states of the standard absolute entropy of organics, namely, the solid state (So(s)), liquid state (So(l)) and gas state (So(g)), were considered in the model. This unified model presented good accuracy and fitting as demonstrated by the R2, F and AARD statistical parameters obtained for 45 So(s) data points, 190 So(l) data points and 174 So(g) data points, respectively. Furthermore, the results of internal and external validation and applicability domain analysis proved that this norm index-based QSPR model was stable and robust and had both statistical quality and predictive ability. The satisfactory obtained results suggested that this unified model can be successfully used to predict the So values of organic compounds in different phase states and that the norm descriptors may be general and universal.

CP-MLR derived QSAR study on the PPARγ agonists: The benzylpyrazole acylsulfonamide derivatives

The PPARγ transactivation activity of benzylpyrazole acylsulfonamide derivatives have been quantitatively analyzed in terms of 0D- to 2D-Dragon descriptors. This study has provided a rational approach for the development of titled derivatives as PPARγ agonists. The descriptors identified in CP-MLR analysis for the PPARγ transactivation activity have highlighted the role of atomic properties (mass, electronegativity, van der Waals volumes and polarizability) in terms of weighted 2D autocorrelations and BCUT descriptors and electronic content in terms of Galvez charge indices and maximal electrotopological positive variation (MAXDP). Additionally, Balaban’s U and centric indices (Uindex and BAC, respectively), Lopping centric index (Lop), topological distance between N and O atom and hydrophobicity accounting parameter MLOGP have also shown prevalence to optimize the PPARγ transactivation of titled compounds. PLS analysis has further confirmed the dominance of the CP‐MLR identified descriptors and applicability domain analysis revealed that the suggested model matches the high quality parameters with good fitting power and the capability of assessing external data and all of the compounds was within the applicability domain of the proposed model and were evaluated correctly.

CP-MLR derived QSAR rationales for the PPARy agonistic activity of the pyridyloxybenzene-acylsulfonamide derivatives

QSAR rationales have been obtained for the PPARy transactivation activity of pyridyloxybenzene-acylsulfonamides in terms of 0D- to 2D-Dragon descriptors. The descriptors identified in CP-MLR analysis have highlighted the role of atomic mass, van der Waals volumes and polarizability through weighted 2D autocorrelations (GATS1v and GATS1p), modified Burden eigenvalue (BEHm4) and molecular weight (MW). Sum of topological distances between O and S atoms (descriptor T(O..S)), and N and Cl atoms (descriptor T(N..Cl)), average connectivity index chi-1(X1A) and Quadratic index (Qindex) have also shown dominance to optimize the PPARγ transactivation. Descriptors RBN and RBF suggested presence of rotatable bonds in a molecular structure for better PPARg activity. Applicability domain analysis revealed that the suggested model matches the high quality parameters with good fitting power and the capability of assessing external data and all of the compounds was within the applicability domain of the proposed model and were evaluated correctly.

QSAR rationales for the PPARα/γ agonistic activity of 4,4-Dimethyl-1,2,3,4-tetrahydroquinoline derivatives

The PPARγ binding affinity and transactivation profiles for hPPARα and hPPARγ of tetrahydroquinoline derivatives have been quantitatively analyzed in terms of topological 0D-, 1D- and 2D-descriptors based on molecular graph theory. Statistically sound models have been obtained between the biological actions and various DRAGON descriptors through combinatorial protocol-multiple linear regression (CP-MLR) computational procedure. Amongst the large number of such derived models, the most significant ones have only been discussed to draw meaningful conclusions. From the statistically significant models, it appeared that the mode of actions of titled compounds were different for hPPARα and hPPARγ transactivation profiles and PPARγ binding affinity. Applicability domain analysis carried out for PPARγ binding affinity revealed that the suggested model matches the high quality parameters with good fitting power and the capability of assessing external data and all of the compounds was within the applicability domain of the proposed model and were evaluated correctly.

Norm index-based QSAR models for acute toxicity of organic compounds toward zebrafish embryo

Zebrafish embryos are highly sensitive to toxicant exposure and have been used to evaluate the potential eco-toxicity caused by organic pollutants in the aquatic environment. This study was to develop four quantitative structure–activity relationship (QSAR) models based on norm descriptors for acute toxicity of different exposure times toward zebrafish embryo of organic compounds with various structures. Norm descriptors were obtained by calculating the norm index of the atomic distribution matrix, which was composed of atomic spatial distribution and atomic properties. These norm index-based QSAR models presented satisfactory results with R2 of 0.8549, 0.9162, 0.8335 and 0.8119 for 48, 96, 120 and 132 h, respectively. Validation results including cross validation, external validation, Y-randomized test and applicability domain analysis indicated that the proposed models were stable, robust and reliable. Accordingly, these norm descriptors might be effective in predicting the acute toxicity of various organics to zebrafish embryos, which might be useful for evaluating the potential hazards of organic pollutants to aquatic environment.

QSAR rationales for the dipeptidyl peptidase-4 (DPP-4) inhibitors: The imidazolopyrimidine amides

The DPP4 inhibition activity of imidazolopyrimidine amides has been quantitatively analyzed in terms of chemometric descriptors. The statistically validated QSAR models provided rationales to explain the inhibition activity of these congeners. The descriptors identified through CP-MLR analysis have highlighted the role of mean electrotopological state (Ms), number of double bonds in molecular structure (nDB), 2D Petitijean shape index (PJI2), Moran autocorrelation of lag-2/weighted by atomic polarizabilities (MATS2p), Moran autocorrelation of lag-6 and lowest eigenvalue n.5 of Burden matrix /weighted by atomic Sanderson electronegativities (MATS6e and BELe5), lowest eigenvalue n.3 and highest eigenvalue n.1 of Burden matrix/weighted by atomic van der Waals volumes (BELv3 and BEHv1). In addition to these 2nd order mean Galvez topological charge index (JGI2), number of ring tertiary C(sp3) (nCrHR) and R--CR--X type structural fragments (C-028) have also shown prevalence to model the inhibitory activity. From statistically validated models, positive contribution of descriptors Ms, PJI2, JGI2, MATS2p, BELe5, BELv3 and BEHv1 suggested that higher values of these are conducive in improving the DPP4 inhibition activity. On the other hand, negative contribution of descriptors nDB, C-028, nCrHR and MATS6e advocated that absence of number of double bonds (nDB), R--CR--X type structural fragment (C-028), number of ring tertiary C(sp3) (nCrHR) and lower value of descriptor MATS6e would be advantageous. PLS analysis has confirmed the dominance of the CP‐MLR identified descriptors and applicability domain analysis revealed the acceptable predictability of suggested models. All the compounds are within the applicability domain of the proposed models and were evaluated correctly.

Prediction of Electrical Conductivity of Deep Eutectic Solvents Using COSMO-RS Sigma Profiles as Molecular Descriptors: A Quantitative Structure–Property Relationship Study

This work presents the development of molecular-based mathematical models for the prediction of electrical conductivity of deep eutectic solvents (DESs). Two new quantitative structure–property relationship (QSPR) models based on conductor-like screening model for real solvent (COSMO-RS) molecular charge density distributions (Sσ-profiles) were developed using the data obtained from the literature. The data comprise 236 experimental electrical conductivity measurements for 21 ammonium- and phosphonium-based DESs, covering a wide range of temperatures and molar ratios. First, the hydrogen-bond acceptors (HBAs) and hydrogen-bond donors (HBDs) of each DES were successfully modeled using COSMO-RS. Then, the calculated Sσ-profiles were used as molecular descriptors. The relation between the conductivity and the descriptors in both models has been expressed via multiple linear regression. The first model accounted for the structure of the HBA, the HBD, the molar ratio, and temperature, whereas the second model additionally incorporated the interactions between the molecular descriptors. The results showed that by accounting for the interactions, the regression coefficient (R2) of the predictive model can be increased from 0.801 to 0.985. Additionally, the scope and reliability of the models were further assessed using the applicability domain analysis. The findings showed that QSPR models based on Sσ-profiles as molecular descriptors are excellent at describing the properties of DESs. Accordingly, the obtained model in this work can be used as a useful guideline in selecting DESs with the desired electrical conductivity for industrial applications.

QSTR Analysis of Acute Rat Oral Toxicity of Amide Pesticides

The rodent acute toxicity is gaining much attention in the ecotoxicological assessment of chemicals. Among the available amide pesticides, the majority of compounds are lacking the experimental toxicity values of rat oral toxicity. In order to explore the structural alerts for toxicity and to fill the toxicity data gap through in silico studies, a series of statistically robust local quantitative structure-toxicity relationship (QSTR) models were developed for the prediction of acute oral toxicity of amide pesticides on rat following OECD principles. The mechanistic interpretation indicated types of amide, the presence of halogen, and SO2 functionality were influential for the toxicity. Applicability domain (AD) analysis and prediction reliability indicators assured the robustness and reliability of the developed models. The detailed analyses of the AD as well the consensus predictions of the unknown compounds were commented for their toxic nature, and prioritization was done for similar classes of compounds without experimental values.

Development of Prediction Models for Drug-Induced Cholestasis, Cirrhosis, Hepatitis, and Steatosis Based on Drug and Drug Metabolite Structures.

Drug-induced liver injury (DILI) is one of the major reasons for termination of drug development. Due to the importance of predicting DILI in early phases of drug development, diverse in silico models have been developed to filter out DILI-causing candidates before clinical study. However, no computational models have achieved sufficient prediction power for screening DILI in early phases because 1) drugs often cause liver injury through reactive metabolites, 2) different clinical outcomes of DILI have different mechanisms, and 3) the DILI label on drugs is not clearly defined. In this study, we developed binary classification models to predict drug-induced cholestasis, cirrhosis, hepatitis, and steatosis based on the structure of drugs and their metabolites. DILI-positive data was obtained from post-market reports of drugs and DILI-negative data from DILIrank, a database curated by the Food and Drug Administration (FDA). Support vector machine (SVM) and random forest (RF) were used in developing models with nine fingerprints and one 2D molecular descriptor calculated from drug (152 DILI-positives and 102 DILI-negatives) and drug metabolite (192 DILI-positives and 126 DILI-negatives) structures. Models were developed according to Organisation for Economic Co-operation and Development (OECD) guidelines for quantitative structure-activity relationship (QSAR) validation. Internal and external validation was performed with a randomization test in order to thoroughly examine model predictability and avoid random correlation between structural features and adverse outcomes. The applicability domain was defined with a leverage method for reliable prediction of new chemicals. The best models for each liver disease were selected based on external validation results from drugs (cholestasis: 70%, cirrhosis: 90%, hepatitis: 83%, and steatosis: 85%) and drug metabolites (cholestasis: 86%, cirrhosis: 88%, hepatitis: 86%, and steatosis: 83%) with applicability domain analysis. Compiled data sets were further exploited to derive privileged substructures that were more frequent in DILI-positive sets compared to DILI-negative sets and in drug metabolite structures compared to drug structures with a Morgan fingerprint level 2.

Application of QSARs in identification of mutagenicity mechanisms of nitro and amino aromatic compounds against Salmonella typhimurium species

In an attempt to describe the underlying causes of mutagenicity mainly due to organic chemicals, quantitative structure-activity relationship (QSAR) models have been developed using two different Salmonella typhimurium mutagenicity endpoints with or without presence of liver metabolic microsomal enzymes (S9) namely TA98-S9 and TA98 + S9. The models were developed using simple 2D variables having definite physicochemical meaning calculated from Dragon, SiRMS, and PaDEL-descriptor software tools. Stepwise regression followed by partial least squares (PLS) regression was used in model development following the strict OECD guidelines for QSAR model development and validation. The models were validated using coefficient of determination R2, cross-validation coefficient Q2LOO (leave one out) while the test set predictions were analyzed using Q2F1 (coefficient of determination for the test set). Several other internationally accepted validation metrics like MAE95%train, average rm(LOO)2 and Δrm(LOO)2 (for the training set) were used to check model robustness while predictive efficiency was evaluated using MAE95%test, average rm2 and Δrm2 (for the test set). The scope of predictions was defined by applicability domain analysis using the DModX approach, a recommended tool for PLS models. The major contributing features related to mutagenicity include lipophilicity, electronegativity, branching and unsaturation, etc. The present manuscript is the first attempt to undertake modeling of two different endpoints (TA98-S9 and TA98 + S9) in order to explore major contributing molecular features linked directly or indirectly to mutagenicity.