- Email: [email protected]
2D QSAR studies of the inhibitory activity of a series of substituted purine derivatives against c-Src tyrosine kinase
Accepted Manuscript Title: 2D QSAR Studies on a Series of substituted purine derivatives inhibitory activity against c-Src tyrosine kinase Author: Mukesh C. Sharma PII: DOI: Reference:
S1658-3655(15)00177-6 http://dx.doi.org/doi:10.1016/j.jtusci.2015.11.002 JTUSCI 257
To appear in: Received date: Revised date: Accepted date:
18-5-2015 10-6-2015 2-11-2015
Please cite this article as: M.C. Sharma, 2D QSAR Studies on a Series of substituted purine derivatives inhibitory activity against c-Src tyrosine kinase, Journal of Taibah University for Science (2015), http://dx.doi.org/10.1016/j.jtusci.2015.11.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
2D QSAR Studies on a Series of substituted purine derivatives inhibitory
ip t
activity against c-Src tyrosine kinase
Mukesh C. Sharma
cr
School of Pharmacy, Devi Ahilya University, Takshila Campus,
us
Khandwa Road, Indore – 452 001, MP, India
Corresponding author:
Ac ce p
te
d
M
an
E-mail: [email protected]
Page 1 of 17
Abstract A Thirty four compounds series of as c-Src tyrosine kinase of substituted purine analogs derivatives were subjected to quantitative structure-activity relationship analyses. Partial least squares regression
ip t
method was applied to derive QSAR models which were further validated for statistical significance by internal and external validation. The best QSAR model developed gave good predictive correlation
cr
coefficient (r2) of 0.8319, significant cross validated correlation coefficient (q2) of 0.7550, r2 for external test set (pred_r2) 0.7983, was developed by PLS method with the descriptors like SsCH3E-
us
index, H-Donor Count, T_2_Cl_3, and negative correlation with SsOHcount. The current study provides better insight into the designing of more potent c-Src tyrosine kinase activity in the future
an
before their synthesis.
M
Keywords: Purine, Quantitative Structure Activity Relationship (QSAR), 2D descriptors, c-Src
Ac ce p
te
d
tyrosine kinase, Partial Least Squares (PLS)
Page 2 of 17
1. Introduction Protein-tyrosine kinases regulate signaling pathways for a broad spectrum of cellular processes including responses to growth factors, neurotransmitters and hormones, activation of the immune
ip t
response, regulation of cell-cell and cell-extracellular matrix interactions, as well as development, oncogenesis, and angiogenesis [1-3]. c-Src kinase is a nonreceptor tyrosine kinase that acts as a signal transduction inhibitor that is a critical component of multiple signaling
cr
pathways that control cell growth, proliferation, invasion, and apoptosis. While c-Src kinase is highly regulated and active only at low levels in most normal cells, studies have shown that c-Src
us
kinase is upregulated in many human tumor types [4-5]. Protein tyrosine kinases (PTKs) are enzymes responsible for the phosphorylation of other proteins and can catalyze the transfer of γ -
an
phosphate group of ATP to the protein phenolic groups (on Tyr). PTKs play a central role in signal transduction pathways and are involved in immune, endocrine, and nervous system
M
physiology and pathology [6]. c-Src tyrosine kinase plays versatile roles in cell responses induced by platelet-derived growth factor (PDGF), that include cell growth, cell cycle progression, cell survival, cell migration, actin cytoskeleton rearrangement, DNA synthesis and
d
receptor endocytosis [7-8]. Src kinases play crucial roles in signal transduction pathways and
te
regulating various cellular functions such as cell proliferation and cell differentiation [9] . The activity and structural conformation of the Src-family protein kinases are mainly regulated by
Ac ce p
phosphorylation events [10]. Src kinase is a protooncogenic tyrosine kinase [11] and has been implicated in the genesis and progression of multiple types of human cancer including colon, breast, lung, and other cancers [12]. Quantitative structure–activity relationship (QSAR) is an attempt to correlate structural or property descriptors of compounds quantitatively with biological activities. The traditional 2D-QSAR model is only a rough approximation to the real relationships as it mainly uses molecular descriptors. QSAR models, mathematical equations relating chemical structure to their biological activity, give information that is useful for drug design and medicinal chemistry [13]. Quantitative structure–activity relationships (QSAR) help to predict the biological activity of new structures and may reveal useful information on structural modification at several substitutional positions of c-Src-binding molecules [14-17]. The present work was undertaken to find a correlation between physicochemical parameters and the biological activity from a series of novel purine derivatives as c-Src tyrosine kinase analogs. The present paper is an attempt for a predictive technique based on the partial least square
Page 3 of 17
regression which identifies key structural features responsible for governing the c-Src tyrosine kinase. These identified important structural features could subsequently be utilized to design novel purine-based anticancer drugs targeting Src tyrosine kinase.
ip t
2. Materials and method 2.1. Computational Methods
cr
The computational studies were performed on a HP window 7 Home Basic running on Intel® V-life MDS (Molecular Design Suite)
TM
us
core processor. The molecular structures of the compounds in the data set were sketched using 3.5 software supplied by V-life Sciences Technologies
[18]. The analogues of purine derivatives reported with potent and selective inhibitory activity
an
against as c-Src tyrosine kinase were taken from the literatures [19]. The biological assay to test the activity of all the molecules was same, and hence, the inhibition values indicated by IC50 are
M
comparable. The biological activities represented by IC50 were converted into the corresponding pIC50 values (-log IC50), which were used as dependent variables in QSAR analysis. For this study, a total of 34 purine derivatives were divided into training and test sets consisting of twenty
te
d
six and eight compounds, (Table 1) respectively. The sphere exclusion method [20] was adopted for division of training and test data set
Ac ce p
comprising of twenty six and eight compounds molecules, respectively, with a dissimilarity value of 8.6, where the dissimilarity value gives the sphere exclusion radius. This algorithm allows constructing training sets covering all descriptor space areas occupied by representative points. Eight compounds, namely 3, 6, 8, 10, 19, 23, 27 and 31 were used as test set, while the remaining molecules were used as the training set. Initially, the data set was split into training (70%) and test sets (30%) using MDS software (care was taken to achieve an even distribution of activities in both the sets (training and test). In order to perform the QSAR analysis, the structures of the compounds in the data set were sketched and the physicochemical descriptors of the molecules were calculated using V-life MDS (molecular design suite) software. All the compounds were batch optimized for the minimization of energies and optimization of geometry using Merck molecular force field, followed by considering distance-dependent dielectric constant of 1.0, convergence criterion or root mean square (RMS) gradient at 0.01 kcal/mol Å and the iteration limit to 10,000 [21].
Page 4 of 17
2.2. Two-dimensional QSAR A large number of theoretical descriptors such as SA Most Hydrophilic (Most hydrophilic value on the vdW surface by Audry Method using Slogp),SA Most Hydrophobic-Hydrophilic Distance
ip t
(distance between most hydrophobic and hydrophilic point on the vdW surface by Audry Method using Slogp), SA Hydrophilic Area (vdW surface descriptor showing hydrophilic surface area by Audry Method using SlogP) and SK Most Hydrophilic (Most hydrophilic value
cr
on the vdW surface by Kellog Method using Slogp), radius of gyration, Wiener’s index, moment of inertia, semi-empirical descriptors, HUMOEnergy: (Highest occupied molecular orbital),
us
heat of formation and ionization potential constitutional, physicochemical, electrostatic, topological and semi-empirical descriptors have been computed from the chemical structures
an
with a view to developing structure activity relationship of purine compounds which would, in turn, predict the biological activity.
M
The preprocessing of the independent variables (i.e. descriptors) was done by removing invariable (constant column), which resulted in a total of 280 descriptors to be used for QSAR
d
analysis. The descriptors having the same value or almost the same value or highly correlated
Ac ce p
2.3. Statistical Computation
te
with other descriptors were removed initially.
To calculate q2, each molecule in the training set was sequentially removed, the model refit using same descriptors, and the biological activity of the removed molecule predicted using the refit model [22]. The q2 was calculated using equation. = 1-
(1)
Where yi, ˆyi are the actual and predicted activity of the ith molecule in the training set, respectively, and ymean is the average activity of all molecules in the training set. For external validation, activity of each molecule in the test set was predicted using the model generated from the training set. The pred_r2 value is calculated as follows (Eq. (2) Pred_r2 =1-
(2)
Page 5 of 17
Where yi, ˆyi are the actual and predicted activity of the ith molecule in the test set, respectively, and ymean is the average activity of all molecules in the training set. Developed quantitative model was evaluated using following statistical measures: N, number of observations (molecules) in the training set; q2, cross-validated r2 (by leave one out) which is
ip t
relative measure of quality of fit; pred_r2, r2 for external test set; q2_se, standard error of crossvalidation and pred_r2 se, standard error of external test set prediction. The low standard error of
cr
pred_r2 se and q2_se shows obsolete quality of fitness of the model. The high pred_r2 and low as deciding factors in selecting the optimal models.
an
3. Results and Discussion
us
pred_r2 se were show high predictive ability of the model. The q2 and pred_r2 values were used
In the present study, training and test sets were generated by using sphere selection method followed by partial least squares regression method. Several 2D QSAR models were constructed,
M
and the best three regression equation obtained is represented:
d
pIC50=1.2458(±0.1665) SsCH3E-index-0.5927(±0.0462) SsOHcount+0.6189(±0.0813) H-Donor
te
Count+0.9188(±0.0651) T_2_Cl_3+0.3085
Degrees of Freedom = 24, Ntraining = 26, Ntest= 8, r2= 0.8319, q2 = 0.7550, F test = 43.148, r2 se =
Ac ce p
0.4572, q2 se = 0.3588, pred_r2 = 0.7983, pred_r2se = 0.2432, Z Score Q^2 = 1.97147, Best Rand Q^2 = 0.58238
QSAR model (1) can explain 79% of the variance in the observed activity values. The r2_se has low value (>0.3) for the regression equation indicating a goodness of fit. The major group of descriptors involved sub groups like SsCH3E-index contribution descriptor, hydrogen donor count and SsOHcount help in understanding the effect of substituent at different position of purine. An estate contribution descriptor SsCH3E-index which represents electro-topological state for number of CH3 group connected with single bond is inversely proportional to the activity of R1 position may lead to an increase in the activity. Molecules (compound 15 and 17) having greater number of methyl groups has more potent Src tyrosine kinase activity. The next contributing descriptor is SsOHcount estate number (~16%), which represents total number of hydroxy group connected with one single bond, show that hydroxyl group should be directly attached with purine ring for maximal determining activity. The alignment descriptor is
Page 6 of 17
T_2_Cl_3 which mean the count of number of double bond separated from chlorine atom by one bond distances in a molecule, indicates that the activity was increase with the presence of chlorine R1 position at purine moiety such as compounds 4, 16 and 28. The most contributing descriptor is H-donor count number of hydrogen bond donor atoms in a molecule, is directly
ip t
proportional to the activity, and suggests that increase in H-donor count of fragment R position is favorable for the activity. The contribution chart of selected descriptors are represented in Fig.
cr
1a and plots of predicted vs. observed values of pIC50 are shown in Fig. 1b. The correlation matrix is shown in Table 2 which shows good correlation of selected parameters with biological
us
activity.
pIC50 = -0.8753(±0.1641) Bromine count +0.1764(±0.0863) SsClE-index+0.2724(±0.0451) mol.
an
wt.
Degrees of Freedom = 24, Ntraining = 26, Ntest= 8, r2 = 0.7311, q2 = 0.6763, F test = 36.416, r2 se =
M
0.4168, q2 se = 0.3921, pred_r2 = 0.7046, pred_r2se = 0.5486, ZScore Q^2 = 2.3657, Best Rand Q^2 = 1.7685.
The another QSAR model-2 using the PLS analysis method and confirmed by the statistical
d
measures viz., squared correlation coefficient (r2), which measures how closely the observed data
te
track the fitted correlation line and standard error (r2_se) which expresses the variation of the residuals or the variation about the regression line. The QSAR model 2 obtained by PLS method
Ac ce p
shows good correlation between biological activity and parameters Bromine count, SsClE-index, and mol.wt the correlation coefficient r2 = 0.7311 and the model explains about 73 % variation in the activity. The model was subjected to cross validation, the value of cross-validated squared correlation coefficient (q2 = 0.6763), low sum of square (pred_r2se = 0.5486) and ZScore Q^2 of predictivity suggested good predictive ability of the biological activity of diversified structure. The descriptor SsClE-index represents electro-topological state indices for number of chlorine atom connected with one single bond, suggests that presence of chlorine atoms in the molecule the higher activity of molecules having chlorine atoms (compounds 4, 16 and 28) that increase in the activity. The descriptor mol.wt. Indices the presence bulky group at R1 position which is detrimental for the activity. The descriptor bromine count (~16%) shows the role of the total number of bromine atom in a molecule. It reveals that presence of electron withdrawing group is favorable for the activity (like compound 5, 17). The plots of predicted vs. observed values of pIC50 are shown in Fig. 1c. Page 7 of 17
pIC50=0.7542(±0.1938) SddsN (nitro) count -0.1685(±0.0763) MostHydrophobic+0.09754 (±0.019) SdsCH Degrees of Freedom = 24, Ntraining = 26, Ntest= 8, r2 = 0.7382, q2 = 0.6732, F test = 20.8543, r2 se = 0.3772, q2 se = 0.4318, pred_r2 = 0.6570, pred_r2se = 0.5218, ZScore Q^2 =1.6542, Best Rand
ip t
Q^2 = 1.1654.
Model (3) shows 73% variance in the observed activity values. The low standard error of r2_se =
cr
0.3772 demonstrates accuracy of the model, parameter for predictivity of test set compounds is high pred_r2 = 0.6570, which shows external predictive power of the model. The descriptor most
us
hydrophobic group inverse relationship of this descriptor suggests that presence of hydrophobic group on purine ring can enhance activity. The descriptor is SddsN (nitro) count suggests the
an
total number of nitro group R1 position will lead to improved activity. The descriptor is SdsCH count indices total number of CH group connected with one double and one single bond. It
M
shows that presence of aliphatic side chain at R position of purine is favorable for the activity. The plots of predicted vs. observed values of pIC50 are shown in Fig. 1d. The above all model is
d
validated by predicting the biological activities of the test molecules, as indicated in Table 3.
te
4. Conclusion:
In the present investigation, QSAR analysis was performed on a data set consist of structurally
Ac ce p
diverse compounds in order to investigate the role of its structural features on their c-Src tyrosine kinase inhibitory activity. The results indicated that the topological, electronic and spatial parameters significantly influence antagonist activity. Electron donating, withdrawing group at R1, and R2 position was found to be an essential feature for c-Src tyrosine kinase. The results obtained from this study indicate the importance of SsCH3E-index, SsOHcount, H-Donor Count , SsClE-index and SddsN(nitro)count in determining the binding affinities of purine analogue for c-Src tyrosine kinase. These investigations will further help in rationalizing the design of c-Src tyrosine kinase molecules.
Acknowledgments The author wishes to express gratitude to V-life Science Technologies Pvt. Ltd for providing the trial version software for the study.
Page 8 of 17
References
ip t
[1] J.S Biscardi, D.A Tice, S. J. Parsons. c-Src, receptor tyrosine kinases, and human cancer,
cr
Adv. Cancer. Res. 76 (1999) 61-119.
[2] S. M. Thomas, J. S. Brugge. Cellular functions regulated by Src family kinases, Annu. Rev.
us
Cell. Dev. Biol. 13, (1997) 513-609.
[3] X. M. Yu, M. W. Salter. Src, a molecular switch governing gain control of synaptic
an
transmission mediated by N-methyl-D-aspartate receptors, Proc. Natl . Acad. Sci .USA 96 (1999) 7697-7704.
[5]
A.Y.Tsygankov,S.K.Shore.
Src:
M
[4] S. V . Russello, S.K. Shore. SRC in human carcinogenesis, Front. Biosci. 9(2004) 139-144. regulation,
role
in
human
carcinogenesis
and
d
pharmacological inhibitors, Curr. Pharm. Des. 10(2004) 1745-1756.
te
[6] T. Hunter. Signalinge 2000 and beyond, Cell 100(2000) 113-127.
Ac ce p
[7] K.A. DeMali, S.L. Godwin, S.P. Soltoff, A. Kazlauskas. Multiple roles for Src in a PDGFstimulated cell. Exp. Cell .Res. 253 (1999) 271–279. [8] S. M. Thomas, J. S. Brugge. Cellular functions regulated by Src family kinases. Annu Rev Cell. Dev. Biol. 13 (1997) 513-609. [9] F.A. Al-Obeidi, J.J. Wu, K.S. Lam. Protein tyrosine kinases: structure, substrate specificity, and drug discovery, Biopolymers 47 (1998) 197-223. [10] Jr R Roskoski. Src kinase regulation by phosphorylation and dephosphorylation. Biochem. Biophys . Res. Comm. 331(2005)1–14. [11] K. Parang, G. Sun. Recent advances in the discovery of Src kinase inhibitors,Expert. Opin. Ther. Patents. 15 (2005) 1183-1207.
Page 9 of 17
[12] R.B. Irby, T.J. Yeatman. Role of Src expression and activation in human cancer, Oncogene 19 (2000) 5636-5642. [13] A. Kurup, R. Grag, D.J. Carini, C. Hansch. Comparative QSAR: angiotensin II antagonists.
ip t
Chem. Rev. 101 (2001) 2727. [14] C. Tintori, M. Magnani, S. Schenone, M. Botta. Docking, 3D-QSAR studies and in silico
cr
ADME prediction on c-Src tyrosine kinase inhibitors. Eur.J.Med. Chem. Mar; 44(2009):990-1000.
us
[15] S.K. Bairy, B.V. Suneel Kumar, J.U. Bhalla, A.B. Pramod, M. Ravikumar, Threedimensional quantitative structure-activity relationship studies on c-Src inhibitors
an
based on different docking methods, Chem. Biol. Drug Des. 73 (2009) 416–427.
Mol. Model. 16 (2010) 361–375.
M
[16] R. Cao, N. Mi, H. Zhang, 3D-QSAR study of c-Src kinase inhibitors based on docking, J.
[17] H. Cao, H. Zhang, X. Zheng, D. Gao, 3D QSAR studies on a series of potent and high
te
245.
d
selective inhibitors for three kinases of RTK family, J. Mol. Graph. Model. 26 (2007) 236–
[18] Vlife MDS software package, version 3.5, supplied by VLife science technologies Pvt. Ltd,
Ac ce p
Pune (2008)
[19] H.Huang, J. Ma, J. Shi, L. Meng, H. Jiang, J. Ding, H Liu. Discovery of novel purine derivatives with potent and selective inhibitory activity against c-Src tyrosine kinase. Bioorg. Med. Chem. 18 (2010)4615-24. [20] A. Golbraikh, A. Tropsha, Predictive QSAR modeling based on diversity sampling of experimental datasets for the training and test set selection, J. Comput. Aided Mol. Des. 16 (2002) 357–369. [21] T.A.Halgren, Merck molecular force field. III. Molecular geome-tries and vibrational frequencies for MMFF94, J. Comput. Chem.17 (1996) 553–586.
Page 10 of 17
[22] R.D. III. Cramer, J.D.Bunce, D.E. Patterson. Cross validation, bootstrapping, and partial least squares compared with multiple regression in conventional QSAR studies. Quant. Struct.
Ac ce p
te
d
M
an
us
cr
ip t
Act. Relat. 7: (1988) 18-25
Page 11 of 17
Table 1: Structure, and biological activity of purine derivatives as c-Src tyrosine kinase
R
R
R
N
R3
R1
R2
1
HN
N
NH2
d
OMe
2
HN
NH2
Compounds 12-23
M
S.No
Compounds 3-11
N
OH
O
HN
N H
us
Compounds 1-2
N
N
N H
N
N
cr
N
N
te
R2
N
N
N
N
NH
an
N
NH
NH
N
ip t
R1
N
N
N N H
N
Compounds 24-34
R
IC50
pIC50
H
2.43
5.614
Set
Training
H
2.42
5.616
Training
H
H
1.21
5.917
Test
4-Chloro
H
H
1.18
5.928
Training
4-Bromo
H
H
1.76
5.754
Training
4-Nitro
H
H
1.4
5.853
Test
3-Nitro
H
H
0.95
6.022
Training
8
4-Aminosulfonyl
H
H
1.82
5.739
Test
9
3-Aminosulfonyl
H
H
0.1
7.000
Training
10
4-Carbamoyl
H
H
2.2
5.657
Test
11
4-Acetamido
H
H
1.05
5.978
Training
4 5 6 7
Ac ce p 3-Methoxy
3
Page 12 of 17
H
H
H
0.75
6.124
Training
13
4-Methoxy
H
H
0.13
6.886
Training
14
3-Methoxy
H
H
0.93
6.031
Training
15
3-Methyl
H
H
0.83
6.08
Training
16
4-Chloro
H
H
0.34
6.468
Training
17
4-Bromo
H
H
0.45
6.346
Training
18
4-Nitro
H
H
0.39
6.408
Training
19
3-Nitro
H
H
0.13
6.886
Test
20
4-Aminosulfonyl
H
H
0.02
7.698
Training
21
3-Aminosulfonyl
H
H
0.12
6.92
Training
22
4-Carbamoyl
H
H
2.01
5.696
Training
23
4-Acetamido
H
H
0.65
6.187
Test
24
H
H
H
3.14
5.503
Training
25
4-Methoxy
H
H
1.33
5.876
Training
26
3-Methoxy
H
H
1.2
5.92
Training
27
3-Methyl
H
H
2.5
5.602
Test
28
4-Chloro
H
H
0.69
6.161
Training
29
4-Nitro
H
H
0.78
6.107
Training
3-Nitro
H
H
0.34
6.468
Training
4-Aminosulfonyl
H
H
0.26
6.585
Test
3-Aminosulfonyl
H
H
0.34
6.468
Training
4-Carbamoyl
H
H
1.22
5.913
Training
4-Acetamido
H
H
1.59
5.798
Training
31 32 33 34
cr
us
an M
d te
Ac ce p
30
ip t
12
Page 13 of 17
0.3276
1.0000
T_2_Cl_3
0.2047
0.5287
SsOHcount
0.4864
0.6188
SsOHcount
cr
H-Donor Count
T_2_Cl_3
us
H-Donor Count
1.0000
0.8064
1.0000
Ac ce p
te
d
M
an
SsCH3E-index
SsCH3Eindex 1.0000
Parameter
ip t
Table. 2. Correlation matrix between descriptors present in the best QSAR model -1
Page 14 of 17
Table 3 Observed and Predicted activities with residue according to 2D QSAR models
2D QSAR Model-1
2D QSAR model-2
1
5.614
Pred. 5.569
Pred. 5.72
Res. -0.106
2 3
5.616 5.917
5.636 5.862
-0.02 0.055
5.599 5.943
0.017 -0.026
4 5
5.928 5.754
5.965 5.649
-0.037 0.105
5.863 5.708
0.065 0.046
5.869 5.622
0.059 0.132
6 7
5.853 6.022
5.914 6.173
-0.061 -0.151
5.821 6.14
0.032 -0.118
5.808 5.941
0.045 0.081
8 9
5.739 7.000
5.691 7.082
0.048 -0.082
5.648 7.119
0.091 -0.119
5.631 6.845
0.108 0.155
10 11
5.657 5.978
5.594 5.888
0.063 0.09
5.579 5.917
0.078 0.061
5.519 5.754
0.138 0.224
12 13
6.124 6.886
6.178 6.814
-0.054 0.072
6.065 6.79
0.059 0.096
6.019 6.797
0.105 0.089
14 15
6.031 6.080
6.195 5.993
-0.164 0.087
5.973 5.932
0.058 0.148
5.931 5.889
0.100 0.191
16 17
6.468 6.346
6.307 6.215
0.161 0.131
6.249 6.142
0.219 0.204
6.247 6.111
0.221 0.235
18 19
6.408 6.886
6.252 6.75
0.156 0.136
6.287 6.622
0.121 0.264
6.316 6.717
0.092 0.169
20 21
7.698 6.920
7.583 6.881
0.115 0.039
7.729 6.825
-0.031 0.095
7.501 6.793
0.197 0.127
22 23
5.696 6.187
5.541 6.014
0.155 0.173
5.765 6.248
-0.069 -0.061
5.569 6.063
0.127 0.124
24 25
5.503 5.876
5.435 5.799
0.068 0.077
5.312 5.727
0.191 0.149
5.404 5.75
0.099 0.126
26 27
5.920 5.602
5.896 5.579
0.024 0.023
2.815 5.549
3.105 0.053
5.792 5.461
0.128 0.141
28 29
6.161 6.107
6.113 6.154
0.048 -0.047
6.107 6.062
0.054 0.045
6.035 6.089
0.126 0.018
30 31
6.468 6.585
6.345 6.495
0.123 0.09
6.251 6.673
0.217 -0.088
6.367 6.363
0.101 0.222
32 33
6.468 5.913
6.424 5.867
0.044 0.046
6.375 5.793
0.093 0.12
6.288 5.769
0.18 0.144
34
5.798
5.695
0.103
5.642
0.156
5.667
0.131
us
an
M
d
Ac ce p
Pred. 5.529
Res. 0.085
5.667 5.822
-0.051 0.095
cr
Res. 0.045
2D QSAR model-3
ip t
pIC50
te
Com
Page 15 of 17
ip t cr us an
Ac ce p
te
d
M
Fig.1a. Plot of contribution chart of 2D QSAR Model
Fig.1b Graphs of observed vs. predicted activity of 2D QSAR model -1
Page 16 of 17
ip t cr us an
Ac ce p
te
d
M
Fig.1c Graphs of observed vs. predicted activity of 2D QSAR model -2
Fig.1d Graphs of observed vs. predicted activity of 2D QSAR model -3
Page 17 of 17
S1658-3655(15)00177-6 http://dx.doi.org/doi:10.1016/j.jtusci.2015.11.002 JTUSCI 257
To appear in: Received date: Revised date: Accepted date:
18-5-2015 10-6-2015 2-11-2015
Please cite this article as: M.C. Sharma, 2D QSAR Studies on a Series of substituted purine derivatives inhibitory activity against c-Src tyrosine kinase, Journal of Taibah University for Science (2015), http://dx.doi.org/10.1016/j.jtusci.2015.11.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
2D QSAR Studies on a Series of substituted purine derivatives inhibitory
ip t
activity against c-Src tyrosine kinase
Mukesh C. Sharma
cr
School of Pharmacy, Devi Ahilya University, Takshila Campus,
us
Khandwa Road, Indore – 452 001, MP, India
Corresponding author:
Ac ce p
te
d
M
an
E-mail: [email protected]
Page 1 of 17
Abstract A Thirty four compounds series of as c-Src tyrosine kinase of substituted purine analogs derivatives were subjected to quantitative structure-activity relationship analyses. Partial least squares regression
ip t
method was applied to derive QSAR models which were further validated for statistical significance by internal and external validation. The best QSAR model developed gave good predictive correlation
cr
coefficient (r2) of 0.8319, significant cross validated correlation coefficient (q2) of 0.7550, r2 for external test set (pred_r2) 0.7983, was developed by PLS method with the descriptors like SsCH3E-
us
index, H-Donor Count, T_2_Cl_3, and negative correlation with SsOHcount. The current study provides better insight into the designing of more potent c-Src tyrosine kinase activity in the future
an
before their synthesis.
M
Keywords: Purine, Quantitative Structure Activity Relationship (QSAR), 2D descriptors, c-Src
Ac ce p
te
d
tyrosine kinase, Partial Least Squares (PLS)
Page 2 of 17
1. Introduction Protein-tyrosine kinases regulate signaling pathways for a broad spectrum of cellular processes including responses to growth factors, neurotransmitters and hormones, activation of the immune
ip t
response, regulation of cell-cell and cell-extracellular matrix interactions, as well as development, oncogenesis, and angiogenesis [1-3]. c-Src kinase is a nonreceptor tyrosine kinase that acts as a signal transduction inhibitor that is a critical component of multiple signaling
cr
pathways that control cell growth, proliferation, invasion, and apoptosis. While c-Src kinase is highly regulated and active only at low levels in most normal cells, studies have shown that c-Src
us
kinase is upregulated in many human tumor types [4-5]. Protein tyrosine kinases (PTKs) are enzymes responsible for the phosphorylation of other proteins and can catalyze the transfer of γ -
an
phosphate group of ATP to the protein phenolic groups (on Tyr). PTKs play a central role in signal transduction pathways and are involved in immune, endocrine, and nervous system
M
physiology and pathology [6]. c-Src tyrosine kinase plays versatile roles in cell responses induced by platelet-derived growth factor (PDGF), that include cell growth, cell cycle progression, cell survival, cell migration, actin cytoskeleton rearrangement, DNA synthesis and
d
receptor endocytosis [7-8]. Src kinases play crucial roles in signal transduction pathways and
te
regulating various cellular functions such as cell proliferation and cell differentiation [9] . The activity and structural conformation of the Src-family protein kinases are mainly regulated by
Ac ce p
phosphorylation events [10]. Src kinase is a protooncogenic tyrosine kinase [11] and has been implicated in the genesis and progression of multiple types of human cancer including colon, breast, lung, and other cancers [12]. Quantitative structure–activity relationship (QSAR) is an attempt to correlate structural or property descriptors of compounds quantitatively with biological activities. The traditional 2D-QSAR model is only a rough approximation to the real relationships as it mainly uses molecular descriptors. QSAR models, mathematical equations relating chemical structure to their biological activity, give information that is useful for drug design and medicinal chemistry [13]. Quantitative structure–activity relationships (QSAR) help to predict the biological activity of new structures and may reveal useful information on structural modification at several substitutional positions of c-Src-binding molecules [14-17]. The present work was undertaken to find a correlation between physicochemical parameters and the biological activity from a series of novel purine derivatives as c-Src tyrosine kinase analogs. The present paper is an attempt for a predictive technique based on the partial least square
Page 3 of 17
regression which identifies key structural features responsible for governing the c-Src tyrosine kinase. These identified important structural features could subsequently be utilized to design novel purine-based anticancer drugs targeting Src tyrosine kinase.
ip t
2. Materials and method 2.1. Computational Methods
cr
The computational studies were performed on a HP window 7 Home Basic running on Intel® V-life MDS (Molecular Design Suite)
TM
us
core processor. The molecular structures of the compounds in the data set were sketched using 3.5 software supplied by V-life Sciences Technologies
[18]. The analogues of purine derivatives reported with potent and selective inhibitory activity
an
against as c-Src tyrosine kinase were taken from the literatures [19]. The biological assay to test the activity of all the molecules was same, and hence, the inhibition values indicated by IC50 are
M
comparable. The biological activities represented by IC50 were converted into the corresponding pIC50 values (-log IC50), which were used as dependent variables in QSAR analysis. For this study, a total of 34 purine derivatives were divided into training and test sets consisting of twenty
te
d
six and eight compounds, (Table 1) respectively. The sphere exclusion method [20] was adopted for division of training and test data set
Ac ce p
comprising of twenty six and eight compounds molecules, respectively, with a dissimilarity value of 8.6, where the dissimilarity value gives the sphere exclusion radius. This algorithm allows constructing training sets covering all descriptor space areas occupied by representative points. Eight compounds, namely 3, 6, 8, 10, 19, 23, 27 and 31 were used as test set, while the remaining molecules were used as the training set. Initially, the data set was split into training (70%) and test sets (30%) using MDS software (care was taken to achieve an even distribution of activities in both the sets (training and test). In order to perform the QSAR analysis, the structures of the compounds in the data set were sketched and the physicochemical descriptors of the molecules were calculated using V-life MDS (molecular design suite) software. All the compounds were batch optimized for the minimization of energies and optimization of geometry using Merck molecular force field, followed by considering distance-dependent dielectric constant of 1.0, convergence criterion or root mean square (RMS) gradient at 0.01 kcal/mol Å and the iteration limit to 10,000 [21].
Page 4 of 17
2.2. Two-dimensional QSAR A large number of theoretical descriptors such as SA Most Hydrophilic (Most hydrophilic value on the vdW surface by Audry Method using Slogp),SA Most Hydrophobic-Hydrophilic Distance
ip t
(distance between most hydrophobic and hydrophilic point on the vdW surface by Audry Method using Slogp), SA Hydrophilic Area (vdW surface descriptor showing hydrophilic surface area by Audry Method using SlogP) and SK Most Hydrophilic (Most hydrophilic value
cr
on the vdW surface by Kellog Method using Slogp), radius of gyration, Wiener’s index, moment of inertia, semi-empirical descriptors, HUMOEnergy: (Highest occupied molecular orbital),
us
heat of formation and ionization potential constitutional, physicochemical, electrostatic, topological and semi-empirical descriptors have been computed from the chemical structures
an
with a view to developing structure activity relationship of purine compounds which would, in turn, predict the biological activity.
M
The preprocessing of the independent variables (i.e. descriptors) was done by removing invariable (constant column), which resulted in a total of 280 descriptors to be used for QSAR
d
analysis. The descriptors having the same value or almost the same value or highly correlated
Ac ce p
2.3. Statistical Computation
te
with other descriptors were removed initially.
To calculate q2, each molecule in the training set was sequentially removed, the model refit using same descriptors, and the biological activity of the removed molecule predicted using the refit model [22]. The q2 was calculated using equation. = 1-
(1)
Where yi, ˆyi are the actual and predicted activity of the ith molecule in the training set, respectively, and ymean is the average activity of all molecules in the training set. For external validation, activity of each molecule in the test set was predicted using the model generated from the training set. The pred_r2 value is calculated as follows (Eq. (2) Pred_r2 =1-
(2)
Page 5 of 17
Where yi, ˆyi are the actual and predicted activity of the ith molecule in the test set, respectively, and ymean is the average activity of all molecules in the training set. Developed quantitative model was evaluated using following statistical measures: N, number of observations (molecules) in the training set; q2, cross-validated r2 (by leave one out) which is
ip t
relative measure of quality of fit; pred_r2, r2 for external test set; q2_se, standard error of crossvalidation and pred_r2 se, standard error of external test set prediction. The low standard error of
cr
pred_r2 se and q2_se shows obsolete quality of fitness of the model. The high pred_r2 and low as deciding factors in selecting the optimal models.
an
3. Results and Discussion
us
pred_r2 se were show high predictive ability of the model. The q2 and pred_r2 values were used
In the present study, training and test sets were generated by using sphere selection method followed by partial least squares regression method. Several 2D QSAR models were constructed,
M
and the best three regression equation obtained is represented:
d
pIC50=1.2458(±0.1665) SsCH3E-index-0.5927(±0.0462) SsOHcount+0.6189(±0.0813) H-Donor
te
Count+0.9188(±0.0651) T_2_Cl_3+0.3085
Degrees of Freedom = 24, Ntraining = 26, Ntest= 8, r2= 0.8319, q2 = 0.7550, F test = 43.148, r2 se =
Ac ce p
0.4572, q2 se = 0.3588, pred_r2 = 0.7983, pred_r2se = 0.2432, Z Score Q^2 = 1.97147, Best Rand Q^2 = 0.58238
QSAR model (1) can explain 79% of the variance in the observed activity values. The r2_se has low value (>0.3) for the regression equation indicating a goodness of fit. The major group of descriptors involved sub groups like SsCH3E-index contribution descriptor, hydrogen donor count and SsOHcount help in understanding the effect of substituent at different position of purine. An estate contribution descriptor SsCH3E-index which represents electro-topological state for number of CH3 group connected with single bond is inversely proportional to the activity of R1 position may lead to an increase in the activity. Molecules (compound 15 and 17) having greater number of methyl groups has more potent Src tyrosine kinase activity. The next contributing descriptor is SsOHcount estate number (~16%), which represents total number of hydroxy group connected with one single bond, show that hydroxyl group should be directly attached with purine ring for maximal determining activity. The alignment descriptor is
Page 6 of 17
T_2_Cl_3 which mean the count of number of double bond separated from chlorine atom by one bond distances in a molecule, indicates that the activity was increase with the presence of chlorine R1 position at purine moiety such as compounds 4, 16 and 28. The most contributing descriptor is H-donor count number of hydrogen bond donor atoms in a molecule, is directly
ip t
proportional to the activity, and suggests that increase in H-donor count of fragment R position is favorable for the activity. The contribution chart of selected descriptors are represented in Fig.
cr
1a and plots of predicted vs. observed values of pIC50 are shown in Fig. 1b. The correlation matrix is shown in Table 2 which shows good correlation of selected parameters with biological
us
activity.
pIC50 = -0.8753(±0.1641) Bromine count +0.1764(±0.0863) SsClE-index+0.2724(±0.0451) mol.
an
wt.
Degrees of Freedom = 24, Ntraining = 26, Ntest= 8, r2 = 0.7311, q2 = 0.6763, F test = 36.416, r2 se =
M
0.4168, q2 se = 0.3921, pred_r2 = 0.7046, pred_r2se = 0.5486, ZScore Q^2 = 2.3657, Best Rand Q^2 = 1.7685.
The another QSAR model-2 using the PLS analysis method and confirmed by the statistical
d
measures viz., squared correlation coefficient (r2), which measures how closely the observed data
te
track the fitted correlation line and standard error (r2_se) which expresses the variation of the residuals or the variation about the regression line. The QSAR model 2 obtained by PLS method
Ac ce p
shows good correlation between biological activity and parameters Bromine count, SsClE-index, and mol.wt the correlation coefficient r2 = 0.7311 and the model explains about 73 % variation in the activity. The model was subjected to cross validation, the value of cross-validated squared correlation coefficient (q2 = 0.6763), low sum of square (pred_r2se = 0.5486) and ZScore Q^2 of predictivity suggested good predictive ability of the biological activity of diversified structure. The descriptor SsClE-index represents electro-topological state indices for number of chlorine atom connected with one single bond, suggests that presence of chlorine atoms in the molecule the higher activity of molecules having chlorine atoms (compounds 4, 16 and 28) that increase in the activity. The descriptor mol.wt. Indices the presence bulky group at R1 position which is detrimental for the activity. The descriptor bromine count (~16%) shows the role of the total number of bromine atom in a molecule. It reveals that presence of electron withdrawing group is favorable for the activity (like compound 5, 17). The plots of predicted vs. observed values of pIC50 are shown in Fig. 1c. Page 7 of 17
pIC50=0.7542(±0.1938) SddsN (nitro) count -0.1685(±0.0763) MostHydrophobic+0.09754 (±0.019) SdsCH Degrees of Freedom = 24, Ntraining = 26, Ntest= 8, r2 = 0.7382, q2 = 0.6732, F test = 20.8543, r2 se = 0.3772, q2 se = 0.4318, pred_r2 = 0.6570, pred_r2se = 0.5218, ZScore Q^2 =1.6542, Best Rand
ip t
Q^2 = 1.1654.
Model (3) shows 73% variance in the observed activity values. The low standard error of r2_se =
cr
0.3772 demonstrates accuracy of the model, parameter for predictivity of test set compounds is high pred_r2 = 0.6570, which shows external predictive power of the model. The descriptor most
us
hydrophobic group inverse relationship of this descriptor suggests that presence of hydrophobic group on purine ring can enhance activity. The descriptor is SddsN (nitro) count suggests the
an
total number of nitro group R1 position will lead to improved activity. The descriptor is SdsCH count indices total number of CH group connected with one double and one single bond. It
M
shows that presence of aliphatic side chain at R position of purine is favorable for the activity. The plots of predicted vs. observed values of pIC50 are shown in Fig. 1d. The above all model is
d
validated by predicting the biological activities of the test molecules, as indicated in Table 3.
te
4. Conclusion:
In the present investigation, QSAR analysis was performed on a data set consist of structurally
Ac ce p
diverse compounds in order to investigate the role of its structural features on their c-Src tyrosine kinase inhibitory activity. The results indicated that the topological, electronic and spatial parameters significantly influence antagonist activity. Electron donating, withdrawing group at R1, and R2 position was found to be an essential feature for c-Src tyrosine kinase. The results obtained from this study indicate the importance of SsCH3E-index, SsOHcount, H-Donor Count , SsClE-index and SddsN(nitro)count in determining the binding affinities of purine analogue for c-Src tyrosine kinase. These investigations will further help in rationalizing the design of c-Src tyrosine kinase molecules.
Acknowledgments The author wishes to express gratitude to V-life Science Technologies Pvt. Ltd for providing the trial version software for the study.
Page 8 of 17
References
ip t
[1] J.S Biscardi, D.A Tice, S. J. Parsons. c-Src, receptor tyrosine kinases, and human cancer,
cr
Adv. Cancer. Res. 76 (1999) 61-119.
[2] S. M. Thomas, J. S. Brugge. Cellular functions regulated by Src family kinases, Annu. Rev.
us
Cell. Dev. Biol. 13, (1997) 513-609.
[3] X. M. Yu, M. W. Salter. Src, a molecular switch governing gain control of synaptic
an
transmission mediated by N-methyl-D-aspartate receptors, Proc. Natl . Acad. Sci .USA 96 (1999) 7697-7704.
[5]
A.Y.Tsygankov,S.K.Shore.
Src:
M
[4] S. V . Russello, S.K. Shore. SRC in human carcinogenesis, Front. Biosci. 9(2004) 139-144. regulation,
role
in
human
carcinogenesis
and
d
pharmacological inhibitors, Curr. Pharm. Des. 10(2004) 1745-1756.
te
[6] T. Hunter. Signalinge 2000 and beyond, Cell 100(2000) 113-127.
Ac ce p
[7] K.A. DeMali, S.L. Godwin, S.P. Soltoff, A. Kazlauskas. Multiple roles for Src in a PDGFstimulated cell. Exp. Cell .Res. 253 (1999) 271–279. [8] S. M. Thomas, J. S. Brugge. Cellular functions regulated by Src family kinases. Annu Rev Cell. Dev. Biol. 13 (1997) 513-609. [9] F.A. Al-Obeidi, J.J. Wu, K.S. Lam. Protein tyrosine kinases: structure, substrate specificity, and drug discovery, Biopolymers 47 (1998) 197-223. [10] Jr R Roskoski. Src kinase regulation by phosphorylation and dephosphorylation. Biochem. Biophys . Res. Comm. 331(2005)1–14. [11] K. Parang, G. Sun. Recent advances in the discovery of Src kinase inhibitors,Expert. Opin. Ther. Patents. 15 (2005) 1183-1207.
Page 9 of 17
[12] R.B. Irby, T.J. Yeatman. Role of Src expression and activation in human cancer, Oncogene 19 (2000) 5636-5642. [13] A. Kurup, R. Grag, D.J. Carini, C. Hansch. Comparative QSAR: angiotensin II antagonists.
ip t
Chem. Rev. 101 (2001) 2727. [14] C. Tintori, M. Magnani, S. Schenone, M. Botta. Docking, 3D-QSAR studies and in silico
cr
ADME prediction on c-Src tyrosine kinase inhibitors. Eur.J.Med. Chem. Mar; 44(2009):990-1000.
us
[15] S.K. Bairy, B.V. Suneel Kumar, J.U. Bhalla, A.B. Pramod, M. Ravikumar, Threedimensional quantitative structure-activity relationship studies on c-Src inhibitors
an
based on different docking methods, Chem. Biol. Drug Des. 73 (2009) 416–427.
Mol. Model. 16 (2010) 361–375.
M
[16] R. Cao, N. Mi, H. Zhang, 3D-QSAR study of c-Src kinase inhibitors based on docking, J.
[17] H. Cao, H. Zhang, X. Zheng, D. Gao, 3D QSAR studies on a series of potent and high
te
245.
d
selective inhibitors for three kinases of RTK family, J. Mol. Graph. Model. 26 (2007) 236–
[18] Vlife MDS software package, version 3.5, supplied by VLife science technologies Pvt. Ltd,
Ac ce p
Pune (2008)
[19] H.Huang, J. Ma, J. Shi, L. Meng, H. Jiang, J. Ding, H Liu. Discovery of novel purine derivatives with potent and selective inhibitory activity against c-Src tyrosine kinase. Bioorg. Med. Chem. 18 (2010)4615-24. [20] A. Golbraikh, A. Tropsha, Predictive QSAR modeling based on diversity sampling of experimental datasets for the training and test set selection, J. Comput. Aided Mol. Des. 16 (2002) 357–369. [21] T.A.Halgren, Merck molecular force field. III. Molecular geome-tries and vibrational frequencies for MMFF94, J. Comput. Chem.17 (1996) 553–586.
Page 10 of 17
[22] R.D. III. Cramer, J.D.Bunce, D.E. Patterson. Cross validation, bootstrapping, and partial least squares compared with multiple regression in conventional QSAR studies. Quant. Struct.
Ac ce p
te
d
M
an
us
cr
ip t
Act. Relat. 7: (1988) 18-25
Page 11 of 17
Table 1: Structure, and biological activity of purine derivatives as c-Src tyrosine kinase
R
R
R
N
R3
R1
R2
1
HN
N
NH2
d
OMe
2
HN
NH2
Compounds 12-23
M
S.No
Compounds 3-11
N
OH
O
HN
N H
us
Compounds 1-2
N
N
N H
N
N
cr
N
N
te
R2
N
N
N
N
NH
an
N
NH
NH
N
ip t
R1
N
N
N N H
N
Compounds 24-34
R
IC50
pIC50
H
2.43
5.614
Set
Training
H
2.42
5.616
Training
H
H
1.21
5.917
Test
4-Chloro
H
H
1.18
5.928
Training
4-Bromo
H
H
1.76
5.754
Training
4-Nitro
H
H
1.4
5.853
Test
3-Nitro
H
H
0.95
6.022
Training
8
4-Aminosulfonyl
H
H
1.82
5.739
Test
9
3-Aminosulfonyl
H
H
0.1
7.000
Training
10
4-Carbamoyl
H
H
2.2
5.657
Test
11
4-Acetamido
H
H
1.05
5.978
Training
4 5 6 7
Ac ce p 3-Methoxy
3
Page 12 of 17
H
H
H
0.75
6.124
Training
13
4-Methoxy
H
H
0.13
6.886
Training
14
3-Methoxy
H
H
0.93
6.031
Training
15
3-Methyl
H
H
0.83
6.08
Training
16
4-Chloro
H
H
0.34
6.468
Training
17
4-Bromo
H
H
0.45
6.346
Training
18
4-Nitro
H
H
0.39
6.408
Training
19
3-Nitro
H
H
0.13
6.886
Test
20
4-Aminosulfonyl
H
H
0.02
7.698
Training
21
3-Aminosulfonyl
H
H
0.12
6.92
Training
22
4-Carbamoyl
H
H
2.01
5.696
Training
23
4-Acetamido
H
H
0.65
6.187
Test
24
H
H
H
3.14
5.503
Training
25
4-Methoxy
H
H
1.33
5.876
Training
26
3-Methoxy
H
H
1.2
5.92
Training
27
3-Methyl
H
H
2.5
5.602
Test
28
4-Chloro
H
H
0.69
6.161
Training
29
4-Nitro
H
H
0.78
6.107
Training
3-Nitro
H
H
0.34
6.468
Training
4-Aminosulfonyl
H
H
0.26
6.585
Test
3-Aminosulfonyl
H
H
0.34
6.468
Training
4-Carbamoyl
H
H
1.22
5.913
Training
4-Acetamido
H
H
1.59
5.798
Training
31 32 33 34
cr
us
an M
d te
Ac ce p
30
ip t
12
Page 13 of 17
0.3276
1.0000
T_2_Cl_3
0.2047
0.5287
SsOHcount
0.4864
0.6188
SsOHcount
cr
H-Donor Count
T_2_Cl_3
us
H-Donor Count
1.0000
0.8064
1.0000
Ac ce p
te
d
M
an
SsCH3E-index
SsCH3Eindex 1.0000
Parameter
ip t
Table. 2. Correlation matrix between descriptors present in the best QSAR model -1
Page 14 of 17
Table 3 Observed and Predicted activities with residue according to 2D QSAR models
2D QSAR Model-1
2D QSAR model-2
1
5.614
Pred. 5.569
Pred. 5.72
Res. -0.106
2 3
5.616 5.917
5.636 5.862
-0.02 0.055
5.599 5.943
0.017 -0.026
4 5
5.928 5.754
5.965 5.649
-0.037 0.105
5.863 5.708
0.065 0.046
5.869 5.622
0.059 0.132
6 7
5.853 6.022
5.914 6.173
-0.061 -0.151
5.821 6.14
0.032 -0.118
5.808 5.941
0.045 0.081
8 9
5.739 7.000
5.691 7.082
0.048 -0.082
5.648 7.119
0.091 -0.119
5.631 6.845
0.108 0.155
10 11
5.657 5.978
5.594 5.888
0.063 0.09
5.579 5.917
0.078 0.061
5.519 5.754
0.138 0.224
12 13
6.124 6.886
6.178 6.814
-0.054 0.072
6.065 6.79
0.059 0.096
6.019 6.797
0.105 0.089
14 15
6.031 6.080
6.195 5.993
-0.164 0.087
5.973 5.932
0.058 0.148
5.931 5.889
0.100 0.191
16 17
6.468 6.346
6.307 6.215
0.161 0.131
6.249 6.142
0.219 0.204
6.247 6.111
0.221 0.235
18 19
6.408 6.886
6.252 6.75
0.156 0.136
6.287 6.622
0.121 0.264
6.316 6.717
0.092 0.169
20 21
7.698 6.920
7.583 6.881
0.115 0.039
7.729 6.825
-0.031 0.095
7.501 6.793
0.197 0.127
22 23
5.696 6.187
5.541 6.014
0.155 0.173
5.765 6.248
-0.069 -0.061
5.569 6.063
0.127 0.124
24 25
5.503 5.876
5.435 5.799
0.068 0.077
5.312 5.727
0.191 0.149
5.404 5.75
0.099 0.126
26 27
5.920 5.602
5.896 5.579
0.024 0.023
2.815 5.549
3.105 0.053
5.792 5.461
0.128 0.141
28 29
6.161 6.107
6.113 6.154
0.048 -0.047
6.107 6.062
0.054 0.045
6.035 6.089
0.126 0.018
30 31
6.468 6.585
6.345 6.495
0.123 0.09
6.251 6.673
0.217 -0.088
6.367 6.363
0.101 0.222
32 33
6.468 5.913
6.424 5.867
0.044 0.046
6.375 5.793
0.093 0.12
6.288 5.769
0.18 0.144
34
5.798
5.695
0.103
5.642
0.156
5.667
0.131
us
an
M
d
Ac ce p
Pred. 5.529
Res. 0.085
5.667 5.822
-0.051 0.095
cr
Res. 0.045
2D QSAR model-3
ip t
pIC50
te
Com
Page 15 of 17
ip t cr us an
Ac ce p
te
d
M
Fig.1a. Plot of contribution chart of 2D QSAR Model
Fig.1b Graphs of observed vs. predicted activity of 2D QSAR model -1
Page 16 of 17
ip t cr us an
Ac ce p
te
d
M
Fig.1c Graphs of observed vs. predicted activity of 2D QSAR model -2
Fig.1d Graphs of observed vs. predicted activity of 2D QSAR model -3
Page 17 of 17