SN Applied Sciences

, 1:1599 | Cite as

QSAR, molecular docking approach on the estrogenic activities of persistent organic pollutants using quantum chemical disruptors

  • Ibrahim Tijjani IbrahimEmail author
  • Adamu Uzairu
  • Balarabe Sagagi
Research Article
Part of the following topical collections:
  1. 1. Chemistry (general)


Quantitative structure–activity relationship (QSAR), for predicting estrogenic activity of persistent organic pollutants (POPS) activity of different compounds used as dataset. Density functional theory using B3LYP/6-31G* quantum chemical calculation method was used to find the optimized geometry of the studied chemical disruptor compounds. Fourteen types of molecular descriptors were used to find out the relation between POPS activity and structural properties. Relevant molecular descriptors were selected by genetic function algorithms. The best model obtained was given a distinct validated, good and robust statistical parameters which include; Square correlation coefficient R2 value of (0.9289), Adjusted determination coefficient, \(R^{2}_{\text{adj}}\) value of (0.9284), leave one out cross validation determination coefficient Q2 value of (0.9548) and external validation as predicted determination coefficient R value of R2 (0.819335). Molecular docking analysis find out, the best lead-compound with the higher negative value score of (− 11.8 kcal/mol) were formed hydrophobic interaction and H-bonding with amino acid residue between the disruptor compounds with their 1 × 7j as receptor. The result obtained from the study is expected to be significant and predict estrogenic activities disruptors of the POPS.


Pops QSAR GFA DFT and molecular docking 

1 Introduction

During the Second World War, scientists have identified some certain chemicals contaminants that exhibit toxic features and are persistent in the environment, bioaccumulative, prone to have long-range atmospheric transboundary migration, deposition, and expected to impose serious health effects on humans, wildlife, and marine biota adjacent to and distant from their origin of discharge. These chemical pollutants are referred to as persistent organic pollutants (POPs) [1]. POPs are usually hydrophobic (water-hating) and lipophilic (fat-loving) chemicals. In marine and terrestrial systems, they come strong to solids, particularly organic matter, evading the aqueous segment. Certainly pops enter the lipids of organisms more easily than the inside of the aqueous medium of cells territory and are stockpiled in fatty tissue. This stockpiling in fatty tissue allows the compounds to persevere in biota, where the metabolism rate is low. Persistent organic pollutants (pops) simultaneously to exist in the surrounding environment for several years, leading serious complications such as learning disabilities, birth defects, cancer, behavioral, neurological issue, reproductive, and immunological disorders in wildlife species and humans [2].

Chemicals are essential component of our daily lives. But some chemicals known as endocrine disruptors, can have harmful and dangerous effect on the body’s endocrine (harmony) system. Hormones act in very small quantity and at precise moment in time to regulate the body’s development, growth, production, metabolism, immunity and behaviors. Endocrine disruptors interfere with natural hormone system and the health effects can be felt long after exposure has stopped. Exposure to endocrine descriptors can have life-long effect and can even have consequences for the next generation to come. Endocrine disruptors chemical (EDCs) and potentials EDC are mostly man-made found, in various material such as pesticide DDT, dioxin and plasticizers such as metals, additives or contaminant in food, and personal care products. EDCs have been subjected to be associated with altered reproductive function in males and females. Increased incidence of breast cancer, abnormality of growth and neurodevelopmental delays in children as well as change in immune system. Human exposure to EDCs occurs via ingestion of food, dust and water inhalation of gases and particle in the air and via the skin. Such Diethystilbestrol (DES) was the first synthetic non-steroidal estrogen which was given between 1940 and 1971 to pregnant women with wrong idea [3]. In 1971, DES was shown [4], to cause clear cell carcinoma a rare vaginal tumor problem occurring in girl and women, therefore USA withdrew used of DES and France withdrew in 1977. A recently research shows that, DES also has a potential to cause a variety of significant adverse medical complication and infertility among DES daughters (Individuals were exposed to DES during their mothers’ pregnancies) [5].

1.1 Action of endocrine discruptors

An endocrine disruptors may mimic or partly mimic natural hormone in the body like estrogen or androgen and also thyroid hormone, if may strict the interaction of natural hormones with their receptors by altering their metabolism in vivo [6], it may also bind to a receptor within a cell and block the endogenous hormone from binding. These chemicals that block or antagonize hormones are antiestrogens and anti-androgens. By snooping with the body’s endocrine system, it give the chance for adverse developmental, reproductive, neurological, an immune effects in both humans. The data set (pops) of this research have divided into natural product, medicine & food additive, PCBS, Phenol, Phthalates, BPA, Polybrominated diphenyl ethers (PBDEs) etc. BPA is an industrial chemical that has been used to make certain plastics and resins since the 1960s. It is found in polycarbonate plastics and are often used in containers that store food and beverages, such as water bottles, and in other consumer goods. Under European legislation (UEL), it has been prohibited use of polycarbonate baby bottles since 2011. Then during a first phase of France’s, BPA was banned in food contact materials effective from January 1, 2013. On January 1, 2015 the second phase of France’s ban on bisphenol A (BPA) became effective. However a recent and systematic review [7] with meta-analysis has found to be, there is a consistent increase in risk of abnormal sperm quality with phthalate, ester group as well as with organochlorine (pops). Early phthalate exposure in pregnant women and also associated [8] with alterations in thyroid hormones leading to autism spectrum disorders and developmental delay. Parabens are esters of p-hydroxybenzoic acid which are widely used as preservatives in cosmetics as well as in foods and drugs. Animal experiments have shown that parabens weakening the estrogenic activity [9]. The use of QSAR approaches in chemical disrupting tests (toxicity) is expected to increase in a variety of applications and to report a number of regulatory tasks [10, 11]. In silico methods can be used to upkeep, prioritization, read-across and screening. Among various in silico approaches, molecular docking, where estrogenic activities disruptors’ (pops) is predicted based on the ligand-receptor complex structure and binding energy, is a promising tool for persistent organic pollutants (disruptor chemicals) screening [12, 13]. Molecular docking strategy is a computational ligand-target docking approach that has been used to evaluate structural complexes of a target receptor with its ligand to realize the chemical and structural basis of a ligand’s target specificity. Molecular docking approach has the potential to be applied for discovering molecular initiating events (MIEs) in the Adverse Outcome Pathway (AOP) framework [14]. Direct binding to nuclear receptor (NRs) is one of the main mechanisms by which EDCs can affect the endocrine system [15]. The interaction between a receptor and its ligand is one of the first reactions in the estrogenic activities disruptors (pops) pathway of chemicals in the AOP framework concept [16].

2 Materials and methods

2.1 Optimization approach and calculations of molecular descriptors

All the molecular structure of estrogenic disruptor compounds were drawn using ChemDraw ultra [17] version 12.02 software and subsequently imported into Spartan 14 wave function software [18] for structural minimization. All the 55 molecules geometry structure of pops were optimized using density functional theory (DFT) method at the B3LYP level of theory and 6-31G* accordingly.

2.2 Model generation

All the structures of 55 Pops compounds were studied by statistical methods based on the Genetic Function Algorithm technique to develop all the models. Number of descriptors in the regression equation is 5, the population and maximum generation are 500 and 1000 correspondingly. The number of scaled LOF smoothness parameter is 0.5, maximum equation length number is 5, and mutation probability is 0.1. GFA algorithm selecting the basic function generally developed good and promising models than those made using stepwise regression approach, the model was assessed using LOF and measured using a slight variation of the unique Friedman formula, this come to chance, the best fitness score can be received. The revised formula of LOF [19] follow
$${\text{LOF}} = {\text{SSE/}}\left( {1 - \frac{C + dp}{M}} \right)^{2}.$$
SSE is the sum of square errors, c is the number of terms in the model, other than the constant terms, d is a user defined smoothing parameter, p is the total number of descriptors enclosed in all model terms (ignoring the constant term) and M is the number of sample in the training set unlike the commonly used least squares measure (LSM), LOF measure cannot constantly be reduced by adding more terms to the regression model (Table 1).

2.3 Quality validations

Internal and external validation parameters were used to develop the consistency and predicted ability of the QSAR model. The QSAR models remained developed using the training set pops compound (optimized by Q2), and then the developed novel compound were validated (externally) using the test set compounds. Multiple linear regression (MLR) was used to display the relationship between the dependent variable Y (ED50) and independent variable X (atomic descriptors). The genetic algorithm-multiple linear regression (GA-MLR) investigation led to the selection of five descriptors that were used to collect a linear model for calculating predictive activity on the persistent organic pollutants (pops). The validation parameters have compared with the minimum recommended values for a generally acceptable QSAR model as shown in Table 2.
Table 2

Minimum recommended value for the evaluation of the quantitative QSAR model







Coefficient of determination

≥ 0.6



Cross validation coefficient

> 0.5



Coefficient of determination for external set

≥ 0.6



Adjusted square correlation coefficient

≤ 0.5


P (95%)

Confidence interval at 95%

≤ 0.05


Next test set

Minimum number of extend test set

≥ 5


R2 − Q2

Difference between R2 and Q2

≤ 0.3

Ravinchandran et al. [20]

2.4 Square of the correlation coefficient (R2)

Describes the friction of the total variation recommended to the model. The closer the value of R2 is to 1.0, the better the regression equation explains in Y variable. R2 is the most normally used internal validation and express as the following:
$${\text{R}}^{2} = 1 - \frac{{\sum \left( {Yobserved - Ypredicted} \right)^{2} }}{{\sum \left( {Yobserved - Ytraining} \right)^{2} }}$$

3 Applicability Domain

Applicability domain (AD) of a QSAR model is the physico-chemical, data on which the training set of the model has been developed and for which it is applicable to make prediction for new lead candidate. The model was authenticated using Williams’s graph, and it’s presented as standardized residuals by the leverages Fig. 1. This method exploited to visualize the applicability domain (AD). Leverage indicates a molecules distance from the centroid of X. [21]. The leverage of molecule in the original space is defined as;
$${\text{h}}_{\text{i}} = {\text{X}}_{\text{i}}^{\text{T}} \left( {{\text{X}}^{\text{T}} {\text{X}}} \right)^{ - 1} {\text{X}}_{\text{i}} .$$
where xi is the descriptor vector of the considered molecule X the descriptor matrix derived from the training set descriptor values.
Fig. 1

Plot of standard versus leverages

The leverage (h*) is defined as:
$${\text{h}}^{*} = \frac{{3\left( {p + 1} \right)}}{n}$$
where n = Number of training set, P = Number of descriptors in a test set.

The above graph (Williams plot, Fig. 1) shows that all the molecules of training set fall within the domain of GFA model (Leverages of h* = 0.43) and five molecules of test set are found to be out of warning leverages, so any molecule found beyond warning leverages are outliers not be consider as novel model. The Williams plot shows blue dot indicating training set molecules while the light-brown dot indicate test set molecules, the graph express high residual model, regular model, bad leverages model and good leverages model but four test set molecules found structural outliers.

4 Results and discussion

4.1 QSAR approach

Five developed QSAR models were obtained and recoded, one out of five models found and marked to be the best model (Model 3) due to statistics significance parameters. The developed model 3 shown (Table 3), both names and symbol descriptors used in the optimization model. And also shown the second (Table 4) for the validation result of the Genetic Function Algorithm of model 3 which was generated from the material studio software. The best QSAR model 3 developed has fulfilled the minimum recommendation value of validation measures for acceptable QSAR model [22].
Table 3

Physicochemical list of descriptors used in the best model



Name of descriptors




Centered Broto-Moreau autocorrelation − lag 3/weighted by van der Waals volumes




Centered Broto-Moreau autocorrelation − lag 2/weighted by Sanderson electronegativities




Geary autocorrelation − lag 2/weighted by charges




Smallest absolute eigenvalue of Burden modified matrix − n 3/weighted by relative first ionization potential




Total information content index (neighborhood symmetry of 1-order)


Table 4

Validation of the genetic function approximation (GFA) from material studio


Equation 1

Equation 2

Equation 3

Equation 4

Equation 5

Friedman LOF












Adjusted R-squared






Cross validated R-squared






Significant Regression






Significance-of-regression F-value






Critical SOR F-value (95%)






Replicate points






Computed experimental error






Lack-of-fit points






Min expt. error for non-significant LOF (95%)






Model 1

IC50 = − 0.001955194 * ATSC3v + 2.361765161 * GATS2c + 3.856551375 * SpMax3_Bhi + 2.141975073* TICi − 0.407294756 * TIC1 + 3.388758, N = 41 R2 = (0.9989), \(R^{2}_{\text{ADJ}}\) = 0.928372, Q2 = 0.944818 and \(R^{2}_{PRED}\) = 0.819335, LOF = 0.005644, Significance-of-regression F-value = 3.34514.

Model 2

IC50 = − 0.001891673 * ATSC3v - + 2.260622850 * ATSC2e+3.868519768 * GATS2c - + 2.387952229 * SpMin3_Bhp- + 2.387952229* TIC1 + 5.376981, N = 41 R2 = (0.912891), \(R^{2}_{\text{ADJ}}\) = 0.938365, Q2 = 0.924845 and \(R^{2}_{PRED}\) = 0.80948, LOF = 0.005667, Significance-of-regression F-value = 1.83E+03.

Model 3

IC50 = − 0.001839735 * ATSC3v + 2.189448988 * ATSC2e+4.153651066 * GATS2c + 2.208946411 * SpMin3_Bhv − 0.070039733 * TIC1 + 5.314197, N = 41 R2 = (0.923891), \(R^{2}_{\text{ADJ}}\) = 0.938365, Q2 = 0.9973 and \(R^{2}_{PRED}\) = 0.80348, LOF = 0.005669, Significance-of-regression F-value = 1.83E+03.

These are the best three equations obtained from material studio software, which generated by training set of the persistent organic pollutant compounds, but equation one (Model 1) found to be the best equation and contribute more than the other descriptors (Table 3) with high value descriptor of (GATS2c + 3.856551375) among the five model generated from the software as internal validation. The descriptor GATS2c in Table 3, described as autocorrelation descriptor lag 2/weighted by charges in the descriptor java class, this explain the contribution of the compound that contains hydrogen oxygen, carbon and nitrogen respectively (Table 5).
Table 5

Best three docking interactions with the estrogenic activities disruptor’s compound, docking scores and active site residues involved


Binding energy

Residual interaction

Hydrogen bond interaction

Hydrogen bond distance


− 11.8

MET336, PHE356, LEU298, ALA302, LEU339, LEU343, LEU298, LEU298, ILE373, ILE376, LEU298, MET336




− 11.7

MET336, PHE356, PHE356, LEU339, LEU343, LEU298, LEU298, LEU476, LEU298, ALA302




− 11.0

MET340, PHE356, LEU298, ALA302, MET336, LEU298, ILE373, PHE356, PHE377, LEU339, LEU343



4.2 Computational docking study

4.2.1 Docking materials

Docking preparation and energy calculation give as (kcal/mol) unit of active pops compound and receptor 1 × 7j were executed by MGL tool and AutoDock Vina of PyRx virtual screening software [23]. Autogrid pre-calculation of the docking estrogenic disruptor molecules was achieved by Autodock Vina of Pyrx by explaining the target point 1 × 7j receptor protein. Energy grid was enrolled based on Lamarckian Genetic Algorithm [24]. Chimera, discovery studio 3.5, Ligplot and PyMol visualization software were used to perform the virtual analysis of (interaction between the molecules and the enzyme) docking site.

4.3 Preparation of the target receptor

The structure of 1 × 7j receptor protein inform of 3D was extracted from the protein data bank in PDB format. All hetero-atomic molecules were excluded from the file using Discovery Studio 3.5 software, hydrogen was added to the receptor and removed water from it by using discovery studio software, the 3D structure of 1 × 7j receptor protein was minimized, protonated and saved in pdbqt file format in all polar residues.

The above 3D and 2D shown the interactions between the ligands and receptor “a and b” (ligand 19-receptor interaction: best docking score), follow by “c and d” and “e and f” both the three ligands and receptor interactions shown the binding site pocket cavity by having H-bond interactions and hydrophobic residue interactions, this result revealed that all the fifty-five persistent organic pollutants (pops) compounds play a significant role as estrogenic activities disruptors compound.

5 Conclusion

The statistically significant model of GFA analysis obtained from material studio software (R2 = (0.923891), \(R^{2}_{\text{ADJ}}\) = 0.938365, Q2 = 0.9973, Friedman LOF = 0.005669) are in good agreement with parameters reported in Table 2, this shows the goodness and reliability of the candidate. A comparison of the predicted toxicities with the experimental log (1/ED50) reported in Table 1 indicated high predictability of the novel candidate evidenced by low residual values observed in the Table, estrogenic disruptor compounds of 55, 36, 26 and 23 are the best predicted in the series evident by its lowest positive residual values. The predicted toxicities of the disruptors compound in log (1/ED50) in Table 1. Shows the promising results with the experimental values (\(R^{2}_{PRED}\) = 0.8348). The plot of experimental toxicities versus predicted log (I/ED50) is shown in Fig. 1 for the test set and Fig. 2 for the training set with the \(R^{2}_{\text{PRED}}\) = 0.8848 also confirm the goodness reliability of the model. The Williams plotted standardized residual versus leverages in the Fig. 2 shows that exploited to visualized the applicability domain AD [25] of training and test set molecule five molecules from test set and two molecules from training set are found to be structural outliers the warning leverages found to be h* = 0.43 in the Fig. 1 Five descriptors ATSC3v, ATSC2e, GATS2c, SpMin3_Bhp and TIC1 generated from the material studio software which are found to be responsible for the estrogenic activities disruptors, similarly GATS2c (Table 3) descriptor found to be more contributor in the estrogenic activities disrupting. Multistep framework of computational docking approach were performed to explore the estrogenic activity disruptors (toxicity) influence of structure features on the estrogenic activities of persistent organic pollutants and investigate the molecular mechanism of receptor (1 × 7js)—ligand (estrogenic activities disruptor compounds) interactions. The detailed binding affinity docking score of estrogenic activity disruptor molecule (Ligand 19. best docking score) were obtained as -11.8 kcal/mol, hydrogen bond distance of 2.16714Å and enclosed by amino acid residues (hydrophobic interactions) of MET336, PHE356, LEU298, ALA302, LEU339, LEU343, ILE373, ILE376, this indicated an acceptable reliability of the parameters specified in the docking approach results (Figs. 3, 4, 5, 6, 7, 8, 9, 10).
Fig. 2

Plot Log1/ED50 of toxicity versus predicted activity

Fig. 3

3D protein receptor 1 × 7j after prepared

Fig. 4

3D prepared best 19 ligand

Fig. 5

3D interaction between receptor 1 × 7j

Fig. 6

2D visualization structure between receptor and Ligand 19 best 1 × 7j and Ligand 19

Fig. 7

3D interaction between receptor 1 × 7j and Ligand 16

Fig. 8

2D visualization structure between 1 × 7j receptor and ligand 16 interactions

Fig. 9

3D Interactions between 1 × 7j receptor and ligand 7

Fig. 10

2D visualization interactions between 1 × 7j receptor and ligand 7

6 Summary

The topic title as “Qsar, molecular docking approach on the estrogenic activities of persistent organic pollutants using quantum chemical disruptors” stated the fifty-five compounds which are persistent organic pollutants against the estrogenic activity, the research paper found out the predicting of the persistent organic pollutants (disruptors of estrogenic activities), based on predicting by using quantum chemical descriptors techniques. The first method quantitative structure activity relationship (QSAR) generate a model among the disruptor compounds of estrogen hormone activities after optimized the whole persistent organic pollutants and generate the descriptors by PaDEL software, this enable to generate the model after dividing the descriptors into training-set and test-set and used a material studio software to generate the model. Similarly second method approach Molecular docking shows the interactions between the receptor enzyme and estrogenic activities disruptor compounds (Pops) using different software of Discovery studio, PyRx software, Ligplus and PyMol software for reviewing the graphics image of ligands and receptor inform of 2D and 3D structure.


Compliance with ethical standard

Conflict of interest

The Authors declared that they have no conflict of interest.


  1. 1.
    Ashraf MA, Sarfraz M, Naureen R, Gharibreza M (2015) Environmental impacts of metallic elements: speciation, bioavailability and remediation. Springer, New YorkCrossRefGoogle Scholar
  2. 2.
    Jones KC, De Voogt P (1999) Persistent organic pollutants (POPs): state of the science. Environ Pollut 100(1–3):209–221CrossRefGoogle Scholar
  3. 3.
    Sweetman AJ, Dalla Valle M, Prevedouros K, Jones KC (2005) The role of soil organic carbon in the global cycling of persistent organic pollutants (POPs): interpreting and modelling field data. Chemosphere 60(7):959–972CrossRefGoogle Scholar
  4. 4.
    Watkins O (1948) Diethylstilbestrol in the prevention and treatment of complications of pregnancy. Am J Obstet Gynecol 56(5):821–834CrossRefGoogle Scholar
  5. 5.
    Hatch E et al (2015) Prenatal diethylstilbestrol exposure and risk of obesity in adult women. J Dev Orig Health Dis 6(3):201–207CrossRefGoogle Scholar
  6. 6.
    Palmer JR et al (2009) Urogenital abnormalities in men exposed to diethylstilbestrol in utero: a cohort study. Environ Health 8(1):37CrossRefGoogle Scholar
  7. 7.
    Wang C et al (2016) The classic EDCs, phthalate esters and organochlorines, in relation to abnormal sperm quality: a systematic review with meta-analysis. Sci Rep 6:19982CrossRefGoogle Scholar
  8. 8.
    Huang P-C, Tsai C-H, Liang W-Y, Li S-S, Huang H-B, Kuo P-L (2016) Early phthalates exposure in pregnant women is associated with alteration of thyroid hormones. PLoS ONE 11(7):e0159398CrossRefGoogle Scholar
  9. 9.
    Philippat C, Bennett DH, Krakowiak P, Rose M, Hwang H-M, Hertz-Picciotto I (2015) Phthalate concentrations in house dust in relation to autism spectrum disorder and developmental delay in the CHildhood Autism Risks from Genetics and the Environment (CHARGE) study. Environ Health 14(1):56CrossRefGoogle Scholar
  10. 10.
    Raies AB, Bajic VB (2016) In silico toxicology: computational methods for the prediction of chemical toxicity. Wiley Interdiscip Rev Comput Mol Sci 6:147–172CrossRefGoogle Scholar
  11. 11.
    Gao Y, Lin Z, Chen R, Wang T, Liu S, Yao Z, Yin D (2012) Using molecular docking to compare toxicity of reactive chemicals to freshwater and marine luminous bacteria. Mol Inform 31:809–816CrossRefGoogle Scholar
  12. 12.
    Rabinowitz JR, Goldsmith MR, Little SB, Pasquinelli MA (2008) Computational Molecular Modeling for evaluating the toxicity of environmental chemicals: prioritizing bioassay requirements. Environ Health Perspect 116:573–576CrossRefGoogle Scholar
  13. 13.
    Al Sharif M, Tsakovska I, Pajeva I, Alov P, Fioravanzo E, Bassan A, Kovarich S, Yang C, Mostrag-Szlichtyng A, Vitcheva V et al (2017) The application of molecular modelling in the safety assessment of chemicals: a case study on ligand-dependent PPARγ dysregulation. Toxicology 392:140–154CrossRefGoogle Scholar
  14. 14.
    Swedenborg E, Rüegg J, Mäkelä S, Pongratz I (2009) Endocrine disruptive chemicals: mechanisms of action and involvement in metabolic disorders. J Mol Endocrinol 43:1–10CrossRefGoogle Scholar
  15. 15.
    Allen TEH, Goodman JM, Gutsell S, Russell PJ (2016) A history of the molecular initiating event. Chem Res Toxicol 29:2060–2070CrossRefGoogle Scholar
  16. 16.
    Hertz-Picciotto I (2015) Phthalate concentrations in house dust in relation to autism spectrum disorder and developmental delay in the CHildhood Autism Risks from Genetics and the Environment (CHARGE) study. Environ Health 14(1):56CrossRefGoogle Scholar
  17. 17.
    Ultra CCD (2001) 7.0, cambridge soft corporation, (property picker activeX control), 100 Cambridge park Dr. Cambridge, MA 02140-2317 USAGoogle Scholar
  18. 18.
    Abdulfatai U, Uzairu A, Uba S, Shallangwa GA (2019) Quantitative structure-properties relationship, molecular dynamic simulations and designs of some novel lubricant additives. Egypt J Petrol 28(2):241–245CrossRefGoogle Scholar
  19. 19.
    Khaled K (2011) Modeling corrosion inhibition of iron in acid medium by genetic function approximation method: a QSAR model. Corros Sci 53(11):3457–3465CrossRefGoogle Scholar
  20. 20.
    Bramhane DM, Kulkarni PA, Martis EA, Pissurlenkar RR, Coutinho EC, Nagarsenker MS (2016) Characterization of pioglitazone cyclodextrin complexes: molecular modeling to in vivo evaluation. J Pharm Bioallied Sci 8(2):161CrossRefGoogle Scholar
  21. 21.
    Monika JK, Singh K (2013) Virtual screening using the ligand ZINC database for novel lipoxygenase-3 inhibitors. Bioinformation 9(11):583CrossRefGoogle Scholar
  22. 22.
    Najafi A, Ardakani SS, Marjani M (2011) Quantitative structure-activity relationship analysis of the anticonvulsant activity of some benzylacetamides based on genetic algorithm-based multiple linear regression. Trop J Pharm Res 10(4):483–490CrossRefGoogle Scholar
  23. 23.
    Vina A (2010) Improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading Trott, Oleg; Olson, Arthur J. J Comput Chem 31:455–461Google Scholar
  24. 24.
    Kumar D, Kumar P, Bhubaneswaran S, Mitra A (2010) Advanced drug designing softwares and their application in medical research. Int J Pharm Sci 2:16–18Google Scholar
  25. 25.
    Olasupo SB, Uzairu A, Sagagi BS (2017) Quantitative structure toxicity relationship (QSTR) models for predicting toxicity of polychlorinated biphenyls (PCBs) using quantum chemical descriptors. Chemistry 2(3):107–117Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of ChemistryKano University of Science and TechnologyKanoNigeria
  2. 2.Department of ChemistryAhmadu Bello University ZariaZariaNigeria

Personalised recommendations