Generated QSAR Model Research Articles

Influence of structure properties on protein-protein interactions-QSAR modeling of changes in diffusion coefficients.

Information about protein-protein interactions provides valuable knowledge about the phase behavior of protein solutions during the biopharmaceutical production process. Up to date it is possible to capture their overall impact by an experimentally determined potential of mean force. For the description of this potential, the second virial coefficient B22, the diffusion interaction parameter kD, the storage modulus G', or the diffusion coefficient D is applied. In silico methods do not only have the potential to predict these parameters, but also to provide deeper understanding of the molecular origin of the protein-protein interactions by correlating the data to the protein's three-dimensional structure. This methodology furthermore allows a lower sample consumption and less experimental effort. Of all in silico methods, QSAR modeling, which correlates the properties of the molecule's structure with the experimental behavior, seems to be particularly suitable for this purpose. To verify this, the study reported here dealt with the determination of a QSAR model for the diffusion coefficient of proteins. This model consisted of diffusion coefficients for six different model proteins at various pH values and NaCl concentrations. The generated QSAR model showed a good correlation between experimental and predicted data with a coefficient of determination R2 = 0.9 and a good predictability for an external test set with R2 = 0.91. The information about the properties affecting protein-protein interactions present in solution was in agreement with experiment and theory. Furthermore, the model was able to give a more detailed picture of the protein properties influencing the diffusion coefficient and the acting protein-protein interactions. Biotechnol. Bioeng. 2017;114: 821-831. © 2016 Wiley Periodicals, Inc.

Pharmacophore modeling, 3D-QSAR, docking study and ADME prediction of acyl 1,3,4-thiadiazole amides and sulfonamides as antitubulin agents

Pharmacophore modeling, molecular docking, and in silico ADME studies have been carried out to determine the binding mode and drug likeliness profile of acyl 1,3,4-thiadiazole amides and sulfonamides as antitubulin agents. A four point pharmacophore model (AAHR.11) was generated using 63 compounds with IC50 values ranging from 3.16 to 505.76μM. A statistically significant 3D-QSAR model was generated from the pharmacophore hypothesis. The model had a high correlation coefficient (R2=0.8925), cross validation coefficient (Q2=0.8204) and F value (44.3) at 6 component PLS factor. The results of external validation indicated that the generated QSAR model possessed a high predictive power (R2=0.83). The generated model also passed Tropsha’s test for predictive ability and Y-Randomisation test. The Domain of Applicability (APD) of the model was also successfully defined to ascertain that a given estimation can be considered reliable. Further, the restrictivity of the model was checked with inactive compounds by enrichment studies using the decoy test. In order to evaluate the effectiveness of the docking protocol, co-crystallized ligand was extracted from the ligand binding domain of the protein and was re-docked into the same position. The conformer obtained on re-docking and the co-crystallized ligand were superimposed and the RMSD between the two was found to be 0.853Å. ADME predictions were also performed for these compounds. Outcomes of the present study have been first utilized to get insight into the molecular feature that promotes bioactivity, and then within screening procedure, have been exploited for the estimation of novel potential antitubulin compounds prior to their synthesis and biological tests.

QSAR modeling to design selective histone deacetylase 8 (HDAC8) inhibitors

HDAC8 inhibitors have become an attractive treatment for cancer. This study aimed to facilitate the identification of potential chemical scaffolds for the selective inhibition of histone deacetylase 8 (HDAC8) using in silico approaches. Non-linear QSAR classification and regression models of HDAC8 inhibitors were developed with support vector machine. Mean impact value-based sequential forward feature selection and grid search strategy were used for molecular descriptor selection and parameter optimization, respectively. The generated QSAR models were validated by leave-one-out cross validation and an external test set. The best QSAR classification model yielded 84% of accuracy on the external test prediction and Matthews correlation coefficient is 0.69. The best QSAR regression model showed low root-mean-square error (0.63) and high squared correlation coefficient (0.53) for the test set. The validated QSAR models together with various drug-like properties, molecular docking and molecular dynamics simulation were sequentially used as a multi-step query in chemical database virtual screening. Finally, two hit compounds were discovered as new structural scaffolds which can be used for further in vitro and in vivo activity analyses. The strategy used in this study could be a promising computational strategy which can be utilized for other target drug design.

QSAR models for antioxidant activity of new coumarin derivatives

This study presents 37 new antioxidant coumarin derivatives and strategies for structural modification to improve their antioxidant activities, the main ferric-reducing antioxidant power (FRAP) assay used to evaluate their antioxidant properties and the generation of validated quantitative structure–activity (antioxidant activity) relationship (QSAR) models. In an attempt to generate QSAR models, structures of all coumarin derivatives in the data set were fully optimized by semi-empirical PM6 method using SPARTAN 10 software. Descriptors were calculated by DRAGON 6.0 software. Multiple linear regression (MLR) models were developed with different training/test set combinations using QSARINS 2.2.1 software. Robustness, reliability and predictive power of the models were tested by internal and external validations. Applicability domain of the best two-descriptor model (nTR = 30; r2 = 0.924; RMSETR = 0.213; nTEST = 7; r2ext = 0.887; RMSEext = 0.255; CCCext = 0.939) was determined. Descriptors appeared in the model revealed that complexity, H-bond donor and lipophilic character are important parameters in describing the antioxidant activity. Apart from the compounds in the data set, we also designed 31 new antioxidant coumarin derivatives and predicted their antioxidant activity using the best two-descriptor model. Most of these compounds are promising antioxidants

Hologram Quantitative Structure Activity Relationship Analysis of JNK Antagonists

c-Jun N-terminal kinase-3 (JNK3) is a member of the mitogen-activated protein kinase family (MAPK), and plays an important role in neurological disorders. Therefore, identification of selective JNK3 inhibitor may contribute towards neuroprotection therapies. In this work, we performed hologram quantitative structure-activity relationship (HQSAR) on a series of thiophene trisubstituted derivatives. The best predictions were obtained for HQSAR model with q2 = 0.628 and r2 = 0.986. Statistical parameters from the generated QSAR models indicated the data is well fitted and have high predictive ability. HQSAR result showed that atom, bond and chirality descriptors play an important role in JNK3 activity and also shows that electronegative groups is highly favourble to enhance the biological activity. Our results could be useful to design novel and selective JNK3 inhibitors.

Computational Analysis of CRTh2 receptor antagonist: A Ligand-based CoMFA and CoMSIA approach

CRTh2 receptor is an important mediator of inflammatory effects and has attracted much attention as a therapeutic target for the treatment of conditions such as asthma, COPD, allergic rhinitis and atopic dermatitis. In pursuit of better CRTh2 receptor antagonist agents, 3D-QSAR studies were performed on a series of 2-(2-(benzylthio)-1H-benzo[d]imidazol-1-yl) acetic acids. There is no crystal structure information available on this protein; hence in this work, ligand-based comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were performed by atom by atom matching alignment using systematic search and simulated annealing methods. The 3D-QSAR models were generated with 10 different combinations of test and training set molecules, since the robustness and predictive ability of the model is very important. We have generated 20 models for CoMFA and 100 models for CoMSIA based on two different alignments. Each model was validated with statistical cut off values such as q2>0.4, r2>0.5 and r2pred>0.5. Based on better q2 and r2pred values, the best predictions were obtained for the CoMFA (model 5 q2=0.488, r2pred=0.732), and CoMSIA (model 45 q2=0.525, r2pred=0.883) from systematic search conformation alignment. The high correlation between the cross-validated/predicted and experimental activities of a test set revealed that the CoMFA and CoMSIA models were robust. Statistical parameters from the generated QSAR models indicated the data is well fitted and have high predictive ability. The generated models suggest that steric, electrostatic, hydrophobic, H-bond donor and acceptor parameters are important for activity. Our study serves as a guide for further experimental investigations on the synthesis of new CRTh2 antagonist.

Comparative Molecular Similarity Indices Analysis of Caspase-3 Inhibitors

Caspases, a family of cysteinyl aspartate-specific proteases plays a central role in the regulation and the execution of apoptotic cell death. Activation of caspases-3 stimulates a signaling pathway that ultimately leads to the death of the cell. Hence, caspase-3 has been proven to be an effective target for reducing the amount of cellular and tissue damage. In this work, comparative molecular similarity indices analysis (CoMSIA) was performed on a series of 3,4-dihydropyrimidoindolones derivatives which are inhibitors of caspase-3. The best predictions were obtained for CoMSIA model (<TEX>$q^2$</TEX> = 0.586, <TEX>$r^2$</TEX> = 0.955). The predictive ability of test set (<TEX>$r^2_{pred}$</TEX>) was 0.723. Statistical parameters from the generated QSAR models indicated the data is well fitted and have high predictive ability. Our theoretical results could be useful to design novel and more potent caspase-3 derivatives.

QSAR analyses of DDT analogues and their in silico validation using molecular docking study against voltage-gated sodium channel of Anopheles funestus

DDT has enjoyed the reputation of a successful pesticide in disease control programme and agricultural practices along with the serious opposition and ban later on due to its biomagnification and toxic action against non-target organisms. The present work was carried out to develop QSAR models for analysing DDT analogues for their pesticidal activity and in silico validation of these models. A 2D-QSAR model was generated using stepwise with multiple regression, and the model with a value of r2 = 0.7324; q2 = 0.6215; pred r2 = 0.7038, containing five descriptors, was selected for further study. The 3D QSAR with CoMFA analysis showed that steric contribution of 21% and electrostatic contribution of about 79% were required for larvicidal activity of DDT analogues. A set of 3430 molecules was generated using the basic DDT skeleton as template, and these were evaluated for their mosquito larvicidal activity using the generated QSAR models and DDT as standard. Eleven molecules were selected for in silico validation of these models. For this, a docking study of the selected molecules against the homology-modelled voltage-gated sodium channel of Anopheles funestus was conducted. The study showed that the activities of these analogues as predicted by 2D-QSAR, 3D-QSAR with CoMFA and dock score were observed to be well correlated.

Comparative Molecular Field Analysis of Caspase-3 Inhibitors

Caspases, a family of cysteinyl aspartate-specific proteases plays a central role in the regulation and the execution of apoptotic cell death. Activation of caspases-3 stimulates a signaling pathway that ultimately leads to the death of the cell. Hence, caspase-3 has been proven to be an effective target for reducing the amount of cellular and tissue damage. In this work, comparative molecular field analysis (CoMFA) was performed on a series of 3, 4-dihydropyrimidoindolones derivatives which are inhibitors of caspase-3. The best predictions were obtained for CoMFA model (<TEX>$q^2=0.676$</TEX>, <TEX>$r^2=0.990$</TEX>). The predictive ability of test set (<TEX>$r^2_{pred}$</TEX>) was 0.688. Statistical parameters from the generated QSAR models indicated the data is well fitted and have high predictive ability. Our theoretical results could be useful to design novel and more potent caspase-3 derivatives.

Docking‐based three‐dimensional quantitative structure–activity relationship (3D‐QSAR) predicts binding affinities to aryl hydrocarbon receptor for polychlorinated dibenzodioxins, dibenzofurans, and biphenyls

Polychlorinated dibenzodioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs) cause toxic effects after binding to an intracellular cytosolic receptor called the aryl hydrocarbon receptor (AhR). Thymic atrophy, weight loss, immunotoxicity, acute lethality, and induction of cytochrome P4501A1 have all been correlated with the binding affinity to AhR. To study the key molecular features for determining binding affinity to AhR, a homology model of AhR ligand-binding domains was developed, a molecular docking approach was employed to obtain docking-based conformations of all molecules in the whole set, and 3-dimensional quantitative structure-activity relationship (3D-QSAR) methodology, namely, comparative molecular field analysis (CoMFA), was applied. A partial least square analysis was performed, and QSAR models were generated for a training set of 59 compounds. The generated QSAR model showed good internal and external statistical reliability, and in a comparison with other reported CoMFA models using different alignment methods, the docking-based CoMFA model showed some advantages.

Alignment-independent QSAR Analysis of SecA Inhibitors

SecA ATPase plays a crucial role in translocation of membrane and secreted polypeptides and proteins in bacteria and therefore a perfect target for novel antimicrobial drug design. Herein, we generated QSAR models with an alignment-independent method. The optimum model obtained for the training set was statistically significant with cross-validation regression coefficient (q2) value of 0.40 and correlation coefficient (r2) value of 0.89. These results suggest that this 3D-QSAR model can be used to guide the development of new SecA inhibitors.

More effective dithiocarbamate derivatives inhibiting carbonic anhydrases, generated by QSAR and computational design

Dithiocarbamates (DTC) are promising compounds with potential applications in antitumoral and glaucoma therapy. Our aim is to understand molecular features affecting DTC interaction with carbonic anhydrases (CAs), zinc-containing enzymes maintaining acid-base balance in blood and other tissues. To this end, we generate QSAR models based on a compound series containing 25 DTC, inhibitors of four human (h) CAs isoforms: hCA I, II, IX and XII. We establish that critical physicochemical parameters for DTC inhibitory activity are: hydrophobic, electronic, steric, topological and shape. The predictive power of our QSAR models is indicated by significant values of statistical coefficients: cross-validated correlation q2 (0.55–0.73), fitted correlation r2 (0.75–0.84) and standard error of prediction (0.47-0.23). Based on the established QSAR equations, we analyse 22 new DTC derivatives and identify DTC dicarboxilic acids derivatives and their esters as potentially improved inhibitors of CA I, II, IX and XII.

Enhancement of P-gylcoprotein modulators of arylmethylamine-phenyl derivatives: an integrative modeling approach

Multidrug resistance (MDR) plays a crucial role in the failure of cancer treatment with chemotherapeutic agents. In order to overcome the MDR, we have developed both 2D and 3D-QSAR models to enhance the structural requirement for P-gp inhibitors. In this work, hologram quantitative structure–activity relationship (HQSAR), comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were performed on a series of arylmethylamine-phenyl derivatives of P-gp inhibitors. The best predictions were obtained for HQSAR model (q 2 = 0.645, r 2 = 0.772), CoMFA model (q 2 = 0.577, r 2 = 0.917), and CoMSIA model (electrostatic, and H-bond donor) (q 2 = 0.610, r 2 = 0.923). Statistical parameters from the generated QSAR models indicated that the data is well fitted and have high predictive ability. HQSAR result showed that donor and acceptor descriptors play an important role in P-gp activity, and methoxy group on the A-ring at R1 position is necessary for increasing activity. The CoMFA and CoMSIA models predicted that steric, electrostatic, and H-bond donor parameters are important for activity toward P-gp. Our theoretical results could be useful to design novel and more potent P-gp derivatives.

3D-QSAR study indicates an enhancing effect of membrane ions on psychiatric drugs targeting serotonin receptor 5-HT1A

Antidepressants and antipsychotics are psychiatric agents used for the treatment of various types of psychiatric diseases. Although currently among the most commonly prescribed drugs, their effectiveness and adverse effects are the topic of many studies and controversial claims. Here we generate QSAR models based on compounds series including 20 drugs recommended for two critical psychiatric diseases: depression and schizophrenia and we use these QSAR models to predict the biological activity of these 20 antidepressants and antipsychotics. We establish the membrane ions' contributions (sodium, potassium, calcium and iron) mediated by water to the antagonism of these drugs at the 5-HT1A receptor. The reliability of our QSAR models in predicting compounds activity is indicated by significant values for cross-validated correlation q² (0.60-0.76) and fitted correlation r² (0.96-0.98) coefficients. Our results indicate that potassium, calcium and iron play a key role for the antagonistic activity of drugs at the 5-HT1A receptor. Moreover, based on the established QSAR equations, we analysed 24 new escitalopram derivatives as possibly improved antidepressants targeting the 5-HT1A receptor. We identified that the presence of methyl groups and hydrogen atoms improves antidepressant activity while the simultaneous presence of ethyl, propyl or halogens decreased drastically antidepressant activity at the 5-HT1A site.

Modeling VEGFR kinase inhibition of aminopyrazolopyridine urea derivatives using topological and physicochemical descriptors: a quantitative structure activity analysis study

A QSAR study has been performed on a series of aminopyrazolopyridine urea derivatives with potent VEGFR kinase inhibitory activity. Structural features responsible for the activity of the compounds were characterized by using physicochemical, topological, and electrotopological descriptors, calculated from the Molecular Design Suite software (V-life MDSTM 3.5). Correlations between the inhibitory activities (KDR and KDRcell) of aminopyrazolopyridine urea derivatives and the calculated descriptors were established through simulated annealing method. The best QSAR models generated from the study explain 89 and 88% of the variation in KDR and KDRcell inhibitory activities, respectively. Internal and external validation methods were used to evaluate the predictive capacity of the generated models. The significant cross-validated correlation coefficient (Q 2 > 0.6) and other statistical parameters suggest that the models exhibited considerable predictivity. The generated QSAR models divulge that factors related to lipophilicity and topological state of atoms in the molecule influences KDR and KDRcell inhibitory activities of aminopyrazolopyridine urea derivatives. A QSAR study has been performed on aminopyrazolopyridine urea derivatives which exhibited VEGFR kinase inhibitory activity, and the results derived show that the lipophilicity and the topological state of atoms in the molecule influences KDR and KDRcell inhibitory activities.

A group center overlap based approach for “3D QSAR” studies on TIBO derivatives

Current challenges in drug designing and lead optimization has reached a bottle neck where the main onus lies on rigorous validation to afford robust and predictive models. In the present study, we have suggested that predictive structure-activity relationship (SAR) models based on robust statistical analyses can serve as effective screening tools for large volume of compounds present either in chemical databases or in virtual libraries. 3D descriptors derived from the similarity-based alignment of molecules with respect to group center overlap from each individual template point and other "alignment averaged," but significant descriptors (ClogP, molar refractivity, connolly accessible area) were used to generate QSAR models. The results indicated that the artificial neural network method (r(2) = 0.902) proved to be superior to the multiple linear regression method (r(2) = 0.810). Cross validation of the models with an external set was reasonably satisfactory. Screening PubChem compound database based on the models obtained, yielded 14 newer modified compounds belonging to the TIBO class of inhibitors, as well as, two novel scaffolds, with enhanced binding efficacy as hits. These hits may be targeted toward potent lead-optimization and help in designing and synthesizing new compounds with potential therapeutic utility.

Quantitative structure–activity relationship studies of a series of non-benzodiazepine structural ligands binding to benzodiazepine receptor

Heuristic method (HM) and radial basis function neural network (RBFNN) methods were proposed to generate QSAR models for a set of non-benzodiazepine ligands at the benzodiazepine receptor (BzR). Descriptors calculated from the molecular structures alone were used to represent the characteristics of the compounds. The six molecular descriptors selected by HM in CODESSA were used as inputs for RBFNN. Compared with the results of HM, more accurate prediction could be obtained from RBFNN. The correlation coefficients ( R) of the nonlinear RBFNN model were 0.9113 and 0.9030 for the training and test sets, respectively. This paper proposed an effective method to design new ligands of BzR based on QSAR.

Structural and chemical basis for enhanced affinity and potency for a large series of estrogen receptor ligands: 2D and 3D QSAR studies

The estrogen receptor (ER) is an important drug target for the development of novel therapeutic agents for the treatment of breast cancer. Progress towards the design of more potent and selective ER modulators requires the optimization of multiple ligand-receptor interactions. Comparative molecular field analyses (CoMFA) and hologram quantitative structure–activity relationships (HQSAR) were conducted on a large set of ERα modulators. Two training sets containing either 127 or 69 compounds were used to generate QSAR models for in vitro binding affinity and potency, respectively. Significant correlation coefficients (affinity models, CoMFA, r 2 = 0.93 and q 2 = 0.79; HQSAR, r 2 = 0.92 and q 2 = 0.71; potency models, CoMFA, r 2 = 0.94 and q 2 = 0.72; HQSAR, r 2 = 0.92 and q 2 = 0.74) were obtained, indicating the potential of the models for untested compounds. The generated models were validated using external test sets, and the predicted values were in good agreement with the experimental results. The final QSAR models as well as the information gathered from 3D contour maps should be useful for the design of novel ERα modulators having improved affinity and potency.

Holographic QSAR of selected esters

The HQSAR (Holographic QSAR) method, which has been recently developed, can offer the ability to rapidly and easily generate QSAR models of high statistical quality and predictive value. HQSAR analysis requires selecting values for parameters that specify the size of the hologram that is to be used, and the size and type of fragment substructures that are to be encoded. The color coding is provided by HQSAR to reflect which molecular fragments may be important contributors to the biological activity. In this work, we studied the quantitative structure activity relationship of selected esters using the HQSAR method. A robust HQSAR model with r 2 (non-cross-validated regression coefficient) of 0.981 and q 2 (cross-validated regression coefficient) of 0.912, was developed after optimizing the fragment size and the hologram length. The color coding analysis, which has rarely been reported before, was done here to explain the outlier successfully.

Investigation of enzyme selectivity in the human CYP2C subfamily: homology modelling of CYP2C8, CYP2C9 and CYP2C19 from the CYP2C5 crystallographic template.

Homology modelling of human CYP2C subfamily enzymes, CYP2C8, CYP2C9 and CYP2C19, based on the rabbit CYP2C5 crystal structure template is reported. The relatively high sequence homologies (75-80%) between the rabbit CYP2C5 and human CYP2C subfamily enzymes tend to indicate that the resulting structures should prove adequate models of these major catalysts of human drug metabolism. Selective substrates of all three human CYP2C enzymes are found to fit closely within the putative active sites in a manner which is consistent with site-directed mutagenesis experiments and known positions of substrate metabolism. The specific interactions between substrates and enzymes can be used to rationalize the variation in substrate binding affinity and generate QSAR models for both inhibition and metabolism via CYP2C family enzymes, yielding a generally good agreement with experimental binding data obtained from Km values, with correlation coefficients (R values) of between 0.97 and 0.99 depending on the QSAR equation produced.