3D QSAR AND MOLECULAR DOCKING STUDIES OF FLAVONOIDS AS SPLEEN TYROSINE KINASE INHIBITORS

3D QSAR AND MOLECULAR DOCKING STUDIES OF FLAVONOIDS AS SPLEEN TYROSINE KINASE INHIBITORS

A. Basu ^{* 1}, A. Sarkar ² and P. Basak ³

Department of Biochemistry ¹, Gurudas College, Kolkata - 700054, West Bengal, India.

Department of Computer Science and Engineering ², School of Bioscience and Engineering ³, Jadavpur University, Kolkata 700032, West Bengal, India.

ABSTRACT: Spleen tyrosine kinase (Syk) is a member of tyrosine kinase family protein. Syk protein plays a vital role during intracellular signal transduction from high affinity IgE receptor (F_cεRI) in allergic reaction. Flavonoids are well known compounds for their anti-allergic properties. In this present work, thirty-four structurally similar flavonoids are investigated as Syk inhibitors by using 3-dimensional quantitative structure-activity relationship (QSAR) models and molecular docking studies. By applying genetic algorithm (GA) method, different types of 3D-MoRSE descriptors are selected from various types of geometrical descriptors for model building. A Multiple Linear Regression is built up, considering the inhibitory effect of seventeen flavonoids on model mast cells by QSARINS software. The model is validated internally using the Q² _LOO criterion in leave-one-out method. William graph and Insubria graph are plotted to identify the application domain for that model. The same model is also externally validated for other seventeen flavonoid compounds. Docking studies of these thirty-four compounds are also performed using AutoDock to identify interacting amino acids of Syk protein. Docking analysis reveals that flavonoid chrysin depicts lowest binding energy, -2.74 with Syk protein. Methoxy substitution causes adverse effect on binding. The flavonoid chrysin is found in carrot. Pharmacokinetics study of chrysin shows that the gastrointestinal absorption of chrysin is high with bioavailability score of 0.55. The current work will help in drug design of human Syk inhibitors and provides information for molecular level of interactions between Syk and the flavonoid group of compounds.

Spleen tyrosine kinase, QSAR, Flavonoids, Allergic reaction, Molecular docking, Chrysin

INTRODUCTION: The spleen tyrosine kinase Syk is a tyrosine kinase protein, which is involved in signal transduction in hematopoietic cells, as well as non-hematopoietic cells.

In first type of cells, it plays crucial roles in normal function of B and T cells, neutrophils, macrophage, aggregation of platelets and cell signalling in mast cells of origin and plays a crucial role in signalling in most of these cells.

It is involved in B and T-cell functions, platelet aggregation, mast cell signalling, neutrophils and macrophages ^{1, 2}. The activity of Syk protein as tumor suppressor is an example of its function in second type of cells.

From this point of view, Syk has been introduced as a target for a new type of anticancer drugs in breast cancer, overian cancer ³. Syk kinase inhibitor is also used in treatment of rheumatoid artharitis ⁴, lupus nephritis ⁵, vascular diseases ⁶, allergic asthma and rhinitis ⁷. When antigens bind to immunoglobulin E (IgE) during allergic reaction, mast cells and basophils are activated to release inflammatory mediators e.g. histamine, prostaglandin, leukotriens, serotonin etc. In both early- and late-phase allergen-induced type I hypersensitivity reactions, Syk protein inhibitor completely blocks serotonin mediator release from mast cell ⁸. Syk binds with FcεRI, the high affinity receptor for IgE, for degranulation in mast cells ¹.  Activation of tyrosine kinases plays the most important role during transmission of downstream signalling events in mast cells. This signal transduction pathway is required for the regulation of mast cell degranulation ⁹.

There are several compounds e.g. triazolopyridine ¹⁰, fostamatinib ¹¹, 6- [(1R, 2S)- 2- Amino- cyclo-hexylamino]- 4- (5, 6-dimethyl-pyridin-2-ylamino)-pyridazine-3-carboxylic acid amide ¹² and a series of pyrazolopyrazine- 3- amine and pyrazolo-pyrimidine-3-amine derivatives ¹³, which are synthesized and screened as Syk protein inhibitors in various immunological processes.

In this work, ethanopharmacological drugs are investigated as potential mast cell stabilizers. Several natural constituents, which are obtained from different herbs, like Holy Basil (Ocimum tenuiflorum) ¹⁴, Chamomile (Matricaria recutita) ¹⁵, kaempferol in Cerbera manghas ¹⁶, grapeseed extract ¹⁷, polyphenols of red wine ¹⁸ are identified as mast cell stabilizers. Different types of chemical compounds are also analysed as natural mast cell stabilizers like flavonoids, stilbenes, curcuminoids ¹⁹. Numerous flavonoids like luteolin, apigenin, quercetin, fisetin, myricetin, morin, naringenin etc. are tested as Syk protein inhibitors during mast cell degranulation reaction ^{20, 21}. Inhibitory effects of those flavonoids along with their structural requirements for antigen - induced degranulation process is estimated ²².

The role of Syk protein in mast cell degranulation reaction will be cleared only when potent, specific inhibitors of Syk are identified and evaluated in clinical trials. To do so we perform Quantitative Structure-Activity Relationship (QSAR) analysis to study the human spleen tyrosine kinase inhibitory activity of a series of flavonoid compounds. The QSAR study is based on the important chemistry principle, which states that similar compounds have similar biological properties ²³. The biological activity of the chemical compounds has correlation with their physicochemical properties, which are known as molecular descriptors ²⁴. Recently QSAR models are developed for new drug designing in cancer therapy ²⁵ and anti-HIV therapy ²⁶. We use the QSARINS software ²⁷ for building a Multiple Linear Regression (MLR) model to recognize a potential flavonoid as Syk protein inhibitor.

In new drug designing study, molecular docking analysis between various ligands and the protein of interest, is gaining huge interest nowadays. In docking study, the binding energy of docked compound helps to determine the strength of interactions between ligand and protein. As an example of ethanopharmacological study, several ligands from Cassia tora, Brassica campestris and Calotropis procera, are targeted against acetylcholinesterase using molecular modelling and docking studies for drug therapy in Alzheimer’s disease ²⁸. In a similar approach we perform the molecular docking study with flavonoid compounds with Syk protein by utilizing AutoDock 4.2.3 molecular docking software ²⁹.

Bioavailability of herbal medicinal product is an assessment of several features e.g. degree of absorption after administration, digestion, elucidation of metabolic pathways (yielding potentially new active compounds), and the assessment of elimination routes and their kinetics ³⁰.  Bioavailability of these natural flavonoids must be considered to determine their pharmacological efficiency as mast cell stabilisers ³¹. This information is very relevant for good understanding of the concentrations that will be achieved in the body after administration of those flavonoids as Syk inhibitor. Using Swiss ADME ^{32, 33}, a web tool we evaluate pharmacokinetics, drug-likeness of flavonoids.

MATERIALS AND METHODS:

Data Set for QSAR Study of Syk Inhibitor: The first step in QSAR model construction is to collect a list of flavonoid compounds with experimentally determined inhibitory activity data. The inhibitory effects of 34 flavoniods on the release of β-hexosaminidase from model human cells are compiled from experimental data ²². Their 2D chemical structures are shown in Fig. 1. The data set of 34 experimental flavonoid compounds is divided into training set for model generation, and a prediction set (marked with * in Table 1) for model validation, both containing 17 compounds. The chemical structures of flavonoids as SMILES format are retrieved from PubChem database ³⁴. The biological activity (IC₅₀) of the molecules are converted to their corresponding pIC₅₀ values, and used as dependent variables in the QSAR calculations Table 1.

FIG. 1: TWO DIMENSIONAL STRUCTURES OF THIRTY - FOUR FLAVONOID COMPOUNDS

TABLE 1: LIST OF FLAVONOIDS WITH THEIR INHIBITORY ACIVITY

Energy Minimization of Ligand Molecules: Energy minimized structure of 34 ligand molecules are obtained using MM2 force field ³⁵ with minimum RMS gradient 0.01.

Molecular Descriptors Calculation: Using energy minimized structures, molecular descriptors are calculated using ChemDes ³⁶.ChemDes is a free web-based platform for the calculation of molecular descriptors and fingerprints, which provide more than 3,679 molecular descriptors that are divided into 61 logical blocks. Among them 3D ChemoPy molecular descriptors ³⁷ e.g. geometric descriptors (12), CPSA descriptors (30), RDF descriptors (180), WHIM descriptors (70), MoRSE descriptors (210) are calculated for each flavonoids.

QSAR Model Construction and Validation: In order to obtain the MLR model, we use the QSARINS software ²⁷ and a genetic algorithm (GA) method is applied to select relevant set of variables from a large number of descriptors (502). Variables are specified with 10 models per size with 0.050 cut off according to QUIK rule ³⁸. The genetic algorithm is set up with population size 200, 2000 generations per size and mutation rate is 20. For selecting the best model with two descriptors, models are arranged by QSARINS, according to R² (the highest value correlates with the good fitness of the model, Q² LOO (the highest value indicates the robustness of the model).

Syk Protein 3D Structure Identification: Here, the X-ray diffraction crystallographic structure of human Spleen Tyrosine Kinase with resolution 1.9Å (EC#: 2.7.10.2) is extracted from Protein Data Bank with PDB ID 4FL3 ³⁹. This protein is prepared as receptor by using the graphical interface of the AutoDock Tools, which is present in the MGL Tools 1.4.6 program ²⁹ by using Chem Office Professional 16 suite. All 34 inhibitors are prepared for docking studies by the same programme in the pbdqt format.

Docking Studies: In docking studies, a 60 X 60 X 60 box with a grid spacing of 2 Å is used. The centroid of the grid map has been set (0, 0, 0) with the coordinates along x, y and z axis. No. of individuals in population is 50, for 1000 generations. The maximum no. of docking is set to 20 with energy evaluations 2500. Docking results are analysed with their binding energies in dlg format by Python Molecule Viewer version 1.5.6 ⁴⁰.

RESULT AND DISCUSSION:

Best QSAR Model Using MLR: By using user friendly QSARINS software ²⁷, we develop different QSAR models with low multi-collinearity between molecular descriptors and good correlation with the biological response. The best MLR model for training set is built up with 3D molecular descriptors namely Molecule Representation of Structures based on Electron Diffraction (MoRSE) and Charged Partial Surface Area descriptor (CPSA). The QSAR model and its statistical parameters are shown below:

QSAR Equation:

log(1/IC₅₀) = - 0.0817 (± 1.4970) MoRSEV5 - 0.0433 (± 0.0179) PNSA3 - 3.2940 (± 0.6482)

Where, MoRSEV5 means Molecule Representation of Structures based on Electron Diffraction signal 05, weighted by atomic van der waals volumes, PNSA3 means atomic charge weighed negative surface area.

N= 17; R²= 0.8737; R²_adj = 0.8557; R²-R²_adj = 0.0180; LOF = 82.8709; K_xx = 0.6603; Delta K = 0.0090; RMSE _tr = 6.9614; MAE _tr = 5.3576; RSS _tr = 823.8344; CCC tr =0.9326; s = 7.6711; F = 48.4212

Here, N is the number of compounds of the training set, R² is the coefficient of determination, R² _adj is adjusted R², s is standard error of estimate, F is variance ratio, LOF is Friedman lack of fit ^{41, 42}, Kxx is the correlation among descriptors ³⁸, Delta K is the difference of the correlation between the descriptors (Kx) and the descriptors plus the responses (Kxy), RMSE_tr is Root Mean Square Error in fitting (for training set), MAE_tr is Mean Absolute Error in fitting (calculated on training set), RSS_tr is Residual Sum of Squares in fitting (also for training set) and CCC_tr is the concordance correlation coefficient calculated over the training set ^{43, 44, 45}. The model projects an R² value is of 0.8737, which means a proper fitness for modelling Syk protein inhibition.

In addition, an R²_adj value of 0.8557, which indicates that it is convenient to add a new descriptor with the model. Similarly, along with the low value of the Kxx (0.6603), we can say that the correlation between the model descriptors is low. Not only that the Delta K parameter has 0.0090 value which is indicative of good correlation between selected descriptors and the inhibitory response, with negligible error on experimental and calculated parameters (RMSE_tr = 6.9614; MAE_tr = 5.3576; s = 7.6711). The scatter plot of experimental versus predicted activity by model equation is shown in Fig. 2.

FIG. 2: SCATTER PLOT OF EXPERIMENTAL IC₅₀ VERSUS PREDICTED VALUES BY QSAR MODEL

FIG. 3: THE VALUES PREDICTED BY LOO VERSUS THE EXPERIMENTAL IC₅₀ VALUES

Model Validation: In internal validation procedure, our model is validated for fitting and stability by using the statistical parameters of cross-validation Leave-one-out (LOO). The value of Q² _LOO is 0.8096 which is comparable with the value of R² (0.8737). The QSAR model is robust considering the values of RMSE_cv = 8.5463, MAE_cv = 6.5591, CCC_cv = 0.8990. In Fig. 3, biological activity values predicted by LOO versus the experimental values are shown (yellow circles for training and blue circle for prediction).

External Validation: In QSAR studies, the predictions of compounds present within the applicability domain (AD) are considered as dependable ⁴⁶. William graph is plotted with leverage and standardized residuals ⁴⁷ in Fig. 4. From this figure, it can be concluded that plot compound 30 and 33 are present outside the limit of critical leverage (h = 0.529). In another approach for detecting the model applicability domain using Insubria plot not only compounds 33 and 30, but also compound 34 are marginally present outside the applicability domain as shown in Fig. 5.

FIG. 4: WILLIAMS PLOT WITH HAT DIAGONAL VALUES VERSUS STANDARDIZED RESIDUALS

FIG. 5: INSUBRIA PLOT WITH APPLICABILITY DOMAIN

Molecular Docking and Visual Representation of Protein - Ligand Binding: Using AutoDock 4.2.3 software, binding energies of all 34 compounds in our dataset are calculated. Analysing the docking results, the compound 3 i.e. flavonoid chrysin (compound 3) is selected with the lowest binding energy (-2.74 kcal/mol) as most suitable flavonoid in Syk inhibition. As shown in Fig. 6, the flavonoid chrysin is present within a binding pocket region, forming hydrogen bonding with Gly 190, Lys 191 and His 209. Hydroxyl (OH) group at position 7 shows stabilizing interaction with basic amino acid residues Lys and His. Other compounds containing 7 OH group e.g. 3', 4', 7 Trihydroxyflavone, Wogonin also shows strong binding interaction with Syk protein. On the contrary, flavonoids substituted with methyl group causes unfavourable effect on protein - ligand binding depicted in their binding energy.

FIG. 6: DOCKING STRUCTURE OF COMPOUND 3, CHRYSIN WITH Syk PROTEIN

Bioavailability Predication: From Swiss ADME ³³, comparing pharmacokinetic data, it can be concluded that chrysin, 3', 4', 7 Trihydroxyflavone and Wogonin are highly absorbed by GI tract. Drug likeness of all these three compounds is positive with zero violation in Liponski rule.

CONCLUSION: In our earlier work, we have identified carboxamidotetrazole derivatives of furan, thiophene, naphthalene and benzothiophene can act as potent mast cell stabilisers on human basophils ⁴⁸. In our present study, by using QSAR model with MLR equation and molecular docking analysis, we want to highlight the anti-allergic effect of flavonoids as human spleen tyrosine kinase) Syk protein inhibitor. Two 3D molecular descriptors e.g. MoRSEV5 and PNSA3 are selected by genetic algorithm (GA).

The first one weights the atomic van der waals volumes of flavonoid compounds. The second one is related with solvent accessible surface area and partial charge of all negatively charged atoms e.g. oxygen atom of hydroxyl groups of flavonoids. From the molecular docking studies of all 34 flavonoids, the interaction between hydroxyl groups in 7 position of flavonoids are identified. 3', 4', 7 Trihydroxyflavone is present in Broad bean pod, ladino clover, Alfalfa and rhizome part of glycyrrhizae radix. Chrysin is present in carrot, alfalfa, parsley and thyme ⁴⁹. These compounds are recognized as ethano-pharmacological drugs as mast cell stabilizers in allergic treatment.

ACKNOWLEDGEMENT: The authors thank Professor Paola Gramatica for QSARINS software support.

CONFLICT OF INTEREST: The authors have no conflict of interest.

REFERENCES:

1. Geahlen RL: Getting Syk: spleen tyrosine kinase as a therapeutic target. Trends in pharmacological sciences. 2014; 35(8): 414-22.
2. Yi YS, Son YJ, Ryou C, Sung GH, Kim JH and Cho JY: Functional roles of Syk in macrophage-mediated inflammatory responses. Mediators of inflammation 2014.
3. Singh A and Singour PK: QSAR studies of pyrazole derivatives containing thiourea as tyrosine kinase inhibitors. Int J Pharm Sci Res 2017; 8(3): 1437-40.doi: 10.13040/IJPSR.0975-8232.8(3).1437-40.
4. Deng GM, Kyttaris VC and Tsokos GC: Targeting Syk in autoimmune rheumatic diseases. Frontiers in immunology 2016; 7.
5. Kitai M, Fukuda N, Ueno T, Endo M, Maruyama T, Abe M, Okada K, Soma M and Matsumoto K: Effects of a spleen tyrosine kinase inhibitor on progression of the lupus nephritis in mice. Journal of Pharmacological Sciences 2017; 134(1): 29-36.
6. Seo HH, Kim SW, Lee CY, Lim KH, Lee J, Choi E, Lim S, Lee S and Hwang KC: A spleen tyrosine kinase inhibitor attenuates the proliferation and migration of vascular smooth muscle cells. Biological research. 2017; 50(1): 1.
7. Ulanova M, Duta F, Puttagunta L, Schreiber AD and Befus AD: Spleen tyrosine kinase (Syk) as a novel target for allergic asthma and rhinitis. Expert opinion on therapeutic targets 2005; 9(5: 901-921.
8. Moy LY, Jia Y, Caniga M, Lieber G, Gil M, Fernandez X and Shin JD: Inhibition of spleen tyrosine kinase attenuates allergen-mediated airway constriction. American journal of respiratory cell and molecular biology 2013; 49(6): 1085-1092.
9. Oettgen HC: Fifty years later: emerging functions of IgE antibodies in host defense, immune regulation, and allergic diseases. Journal of Allergy and Clinical Immunology. 2016; 137(6): 1631-45.
10. Ferguson GD, Delgado M, Plantevin-Krenitsky V, Jensen-Pergakes K, Bates RJ, Torres S, Celeridad M, Brown H, Burnett K, Nadolny L and Tehrani L: A novel Triazolopyridine-based spleen tyrosine kinase inhibitor that arrests joint inflammation. PloS one 2016; 11(1): e0145705.
11. Kunwar S, Devkota AR and Ghimire DK: Fostamatinib, an oral spleen tyrosine kinase inhibitor, in the treatment of rheumatoid arthritis: a meta-analysis of randomized controlled trials. Rheumatology international 2016; 36(8): 1077-87.
12. Liao C, Hsu J, Kim Y, Hu DQ, Xu D, Zhang J, Pashine A, Menke J, Whittard T, Romero N and Truitt T: Selective inhibition of spleen tyrosine kinase (SYK) with a novel orally bioavailable small molecule inhibitor, RO9021, impinges on various innate and adaptive immune responses: implications for SYK inhibitors in autoimmune disease therapy. Arthritis research and therapy 2013; 15(5): R146.
13. Shen J, Li X, Zhang Z, Luo J, Long H, Tu Z, Zhou X, Ding K and Lu X: 3-aminopyrazolopyrazine derivatives as spleen tyrosine kinase inhibitors. Chemical biology & drug design 2016; 88(5): 690-8.
14. Finn DF and Walsh JJ: Twenty-first century mast cell stabilizers. British journal of pharmacology 2013; 170(1): 23-37.
15. Zhang T, Finn DF, Barlow JW and Walsh JJ: Mast cell stabilisers. European journal of pharmacology 2016; 778: 158-68.
16. Jeong HY, Sung GH, Kim JH, Yoon JY, Yang Y, Park JG, Kim SH, Yi YS, Yang WS, Yoon DH and Kim TW: Syk and Src are major pharmacological targets of a Cerbera manghas methanol extract with kaempferol-based anti-inflammatory activity. Journal of Ethnopharmacology 2014; 151(2): 960-9.
17. Chen BH, Hung MH, Chen JY, Chang HW, Yu ML, Wan L, Tsai FJ, Wang TP, Fu TF and Chiu CC: Anti-allergic activity of grapeseed extract (GSE) on RBL-2H3 mast cells. Food chemistry 2012; 132(2): 968-74.
18. Calabriso N, Scoditti E, Massaro M, Pellegrino M, Storelli C, Ingrosso I, Giovinazzo G and Carluccio MA: Multiple anti-inflammatory and anti-atherosclerotic properties of red wine polyphenolic extracts: differential role of hydroxycinnamic acids, flavonols and stilbenes on endothelial inflammatory gene expression. European journal of nutrition 2016; 55(2): 477-89.
19. Matsuda H, Nakamura S and Yoshikawa M: Degranulation inhibitors from medicinal plants in antigen-stimulated rat basophilic leukemia (RBL-2H3) cells. Chemical and Pharmaceutical Bulletin 2016; 64(2): 96-103.
20. Shichijo M, Yamamoto N, Tsujishita H, Kimata M, Nagai H and Kokubo T: Inhibition of syk activity and degranulation of human mast cells by flavonoids. Biological and Pharmaceutical Bulletin 2003; 26(12): 1685-1690.
21. Kim SH, Park JG, Lee J, Yang WS, Park GW, Kim HG, Yi YS, Baek KS, Sung NY, Hossen MJ and Lee MN: The dietary flavonoid Kaempferol mediates anti-inflammatory responses via the Src, Syk, IRAK1, and IRAK4 molecular targets. Mediators of inflammation 2015, doi:10.1155/ 2015/904142
22. Mastuda H, Morikawa T, Ueda K, Managi H and Yoshikawa M: Structural requirements of flavonoids for inhibition of antigen-induced degranulation, TNF-α and IL-4 production from RBL-2H3 cells. Bioorganic & medicinal chemistry 2002; 10(10): 3123-3128.
23. Kandemirli F: Introductory Chapter: Some Quantitative Structure Activity Relationship Descriptor. In Quantitative Structure-activity Relationship 2017. In Tech.
24. Leeson PD: Molecular inflation, attrition and the rule of five. Advanced drug delivery reviews 2016; 101: 22-33.
25. Dubey S and Bhosle P: Quantitative Structure Activity Relationship of 2, 5, 6–Trisubstituted imidazo (2, 1-b)-1, 3, 4–Thiadiazole as Anticancer Compounds. Indian Journal of Pharmaceutical Education and Research 2016; 50(1): 198-204.
26. Patil S, Asgaonkar K, Chitre T, Bhat V, Ethape S, Sujalegaonkar A, Bhalekar S: 2D and 3D QSAR of Benzimidazole Analogues as Novel HIV-1 Non Nucleoside Reverse Transcriptase Inhibitors. Indian Journal of Pharmaceutical Education and Research 2017; 51(2): 122-128.
27. Gramatica P, Chirico N, Papa E, Cassani S and Kovarich S: QSARINS: New software for the development, analysis, and validation of QSAR MLR models. Journal of Computational Chemistry 2013; 34(24): 2121-32.
28. Rohit M, Ashok T, Vijaykumar R and Kashniyal K: Molecular Docking Study of Cassia tora, Brassica campestris and Calotropis procera as Acetylcholinesterase Inhibitor. Indian Journal of Pharmaceutical Education and Research 2016; 50(1): 116-122.
29. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew R K, Goodsell DS and Olson AJ: Autodock4 and AutoDock Tools4: automated docking with selective receptor flexiblity. Computational Chemistry 2009; 16: 2785-91.
30. Thelingwani R and Masimirembwa C: Evaluation of herbal medicines: value addition to traditional medicines through metabolism, pharmacokinetic and safety studies. Current drug metabolism 2014; 15(10): 942-52.
31. Derendorf H: Pharmacokinetics of Natural Compounds. Planta Medica 2012; 78(11): IL41.
32. Swiss ADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci. Rep. 2017; 7: 42717.
33. Daina A, Michielin O and Zoete V: Swiss ADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Scientific Reports 2017; 7: 42717.
34. Kim S, Thiessen PA, Bolton EE, Chen J, Fu G, Gindulyte A, Han L, He J, He S, Shoemaker BA, Wang J, Yu B, Zhang J and Bryant SH: PubChem Substance and Compound databases. Nucleic Acids Res. 2016; 44(D1): D1202-13. Epub 2015 [PubMed PMID: 26400175]doi: 10.1093/nar/gkv951.
35. Allinger NL, Yuh YH and Lii JH: Molecular mechanics. The MM3 force field for hydrocarbons.Journal of the American Chemical Society 1989; 111(23): 8551-8566.
36. Dong J, Cao DS, Miao HY, Liu S, Deng BC, Yun YH, and Chen AF: ChemDes: An integrated web-based platform for molecular descriptor and fingerprint computation Journal of cheminformatics 2015; 7(1): 60.
37. Cao DS, Xu QS, Hu QN and Liang YZ: ChemoPy: Freely available python package for computational biology and chemoinformatics Bioinformatics 2013; btt105.
38. Chirico N and Gramatica: Real external predictivity of QSAR models. Part 2. New intercomparable thresholds for different validation criteria and the need for scatter plot inspection. Journal of Chemical Information and Modeling 2012; 52(8): 2044-58.
39. Grädler U, Schwarz D, Dresing V, Musil D, Bomke J, Frech M and Wegener A: Structural and biophysical characterization of the Syk activation switch. Journal of molecular biology 2013; 425(2): 309-333.
40. Sanner MF: Python: a programming language for software integration and development. J. Mol Graph Model 1999; 17(1): 57-61.
41. Friedman JH: Multivariate adaptive regression splines. The annals of statistics 1991; 1-67.
42. Rácz A, Bajusz D and Héberger K: Consistency of QSAR models: Correct split of training and test sets, ranking of models and performance parameters. SAR and QSAR in Environmental Research 2015; 26(7-9): 683-700.
43. Chirico N and Gramatica P: Real external predictivity of QSAR models: how to evaluate it? Comparison of different validation criteria and proposal of using the concordance correlation coefficient. Journal of chemical information and modelling 2011; 51(9): 2320-35.
44. Gramatica P: Principles of QSAR models validation: internal and external. Molecular Informatics 2007; 26(5): 694-701.
45. Gramatica P and Sangion A: A historical excursus on the statistical validation parameters for QSAR models: A clarification concerning metrics and terminology. Journal of chemical information and modelling 2016; 56(6): 1127-31.
46. Sahigara F, Mansouri K, Ballabio D, Mauri A, Consonni V and Todeschini R: Comparison of different approaches to define the applicability domain of QSAR models. Molecules 2012; 17(5): 4791-810.
47. Gramatica P: External Evaluation of QSAR Models, in Addition to Cross Validation: Verification of Predictive Capability on Totally New Chemicals. Molecular informatics 2014; 33(4): 311-4.
48. Basu A, Sarkar A and Basak P: QSAR Model for Mast Cell Stabilizing Activity of Indolecarboxamidotetrazole Compounds on Human Basophils. In: Mandal J, Dutta P, Mukhopadhyay S. (eds) Computational Intelligence, Communications, and Business Analytics. CICBA 2017. Communications in Computer and Information Science 2017; 776: 17-29
49. Upadhyaya K, Saini B and Ashok PK: a review on potential dietary health benefit of flavonoids. International Journal of Chemistry, Pharmacy & Technology 2017; 2(1): 25-33.
    

    ---

    How to cite this article:

    Basu A, Sarkar A and Basak P: 3D QSAR and molecular docking studies of flavonoids as spleen tyrosine kinase inhibitors. Int J Pharm Sci Res 2018; 9(5): 1921-28.doi: 10.13040/IJPSR.0975-8232.9(5).1921-28.

    ---

    All © 2013 are reserved by International Journal of Pharmaceutical Sciences and Research. This Journal licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.

    ---