Research Journal of Chemical Sciences ______________________________________________ ISSN 2231-606X Vol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 47 Valence Connectivity Indices and Shape Indices Based Study of Testosterone Derivatives as SHBG LigandSingh R.K.* and Khan Mohd. Adil Department of Chemistry, M.L.K. P.G. College, Balrampur, UP, INDIAAvailable online at: www.isca.in Received 22nd March 2013, revised 31st March 2013, accepted 6th April 2013Abstract Six descriptors namely shape indices (of order 1, 2 and 3) and valence connectivity indices (of order 0, 1 and 2) of forty SHBG binding testosterone derivatives have been calculated with the help of CAChe Pro of Fujitsu software using the semiemperical PM3 Hamiltonian. Observed biological activities of all forty compounds are in terms of pKd (SHBG-ligand dissociation parameter). These six descriptors have been used in deriving regression models. Shape indices (of order 1, 2 and 3) appear good descriptors for QSAR study of testosterone derivatives. Among these three descriptors, shape index (of order 3) is the best. The top five QSAR models developed from these descriptors have high predictive power and can be used to find out the SHBG activity of any new derivative of testosterone. The quality of regression has been adjudged by correlation coefficient, cross validation coefficient and statistical parameters obtained from Smith’s Statistical Package (version 2.80). Keywords: Testosterone, SHBG, valence connectivity indices, shape indices. IntroductionThe sex steroid testosterone controls various aspects of sexual differentiation, gonadal development, and the growth and functional maturation of reproductive tissues. Sex steroids or their immediate precursors are produced in the steroidogenic cells of the gonads, adrenal glands and placenta, and are transported in the blood to their target tissues by several steroid-binding proteins. A plasma glycoprotein, known as sex hormone-binding globulin (SHBG), binds biologically active androgens and is found in the blood. Because of its very high ligand binding affinity, plasma SHBG is the major plasma transport protein for biologically active androgens and estrogens. It is synthesized by liver cells and has a 7-day half life in circulation. Changes in the blood levels of SHBG influence their plasma distribution and access to target tissues and cells. SHBG greatly influences the bioavailable testosterone level because it binds with high affinity to a large fraction of the testosterone in circulation. The most preferred ligand of SHBG is the androgen 5-dihydrotestosterone. SHBG was shown to mediate hormone action by a hormone-SHBG membrane-receptor complex by the generation of the second messengercAMP inside the cells. It has also been found that numerous human diseases such as endometrial cancer, ovarian dysfunction, male and female infertility, osteoporosis, diabetes, and cardiovascular diseases are associated with abnormal SHBG levels, implicating the protein as a prospective drug target6-8. QSAR study of SHBG binding compounds has successfully done by Cherkasov9,10. A set of parameters called inductive descriptors, which quantify the inductive effect of the electronegative atoms in a molecule, has been used by them. Also such computational methods gained much importance and recently have been applied to different compounds11-15. In this paper, QSAR study of forty SHBG binding testosterone derivatives has been done with the help of shape indices of order 1, 2 and 316,17 and valence connectivity indices of order 0, 1 and 218,19. Observed biological activities of all forty compounds are in terms of pKd (SHBG-ligand dissociation parameter) which is different from earlier studies20-22. Theory: The values of six descriptors have been derived by solving the relevant equations given below: (i) Valence connectivity index (): This index, originally defined by Randic and subsequently refined by Kier and Hall, is a series of numbers designated by "order" and "subgraph type" 23, 24. There are four subgraph types; path, cluster, path/cluster, and chain. These types emphasize different aspects of atom connectivity within a molecule, the amount of branching, ring structures present and flexibility. It is calculated from the hydrogen suppressed molecular graph and defined as follows, 1/2 1111SmNvikkmvd+==  =   Where, () (1) kkkkZHZZ -- - valence connectivity for the k-th atom in the molecular graph, Z= the total number of electrons in the k-th atom, Zk = the number of valence electrons in the k-th atom, H= the number of hydrogen atoms directly attached to the kth non-hydrogen atom, m = 0 - atomic valence connectivity indices (called order-0), m = 1 - one bond path valence connectivity indices (called order-1), m = 2 - two bond fragment valence connectivity indices (called order-2). Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 48 (ii) Shape index (): These indices compare the molecule graph with “minimal” and “maximal” graphs, where the meaning of “minimal” and “maximal” depends on the order “n”. This is intended to capture different aspects of the molecular shape. Kier was first to propose shape indices for molecular graphs, the so called kappa shape indices. The first order kappa shape index (1 or ) is given by, Where, iP = Length of paths of bond length i in the hydrogen suppressed molecule and A is the number of non hydrogen atoms in the molecule. The second order kappa shape index (2 or ) is given by The third order kappa shape index (3 or ) is given by Material and Methods Forty derivatives of testosterone that have been taken from literature are used as study material. These are listed in table-1 alongwith their observed biological activity in terms of SHBG-ligand dissociation parameter i.e. pKd. The parent skeleton structure of testosterone is shown in figure-1. The QSAR study of all the derivatives of testosterone has been made with the help of six descriptors given below; i. Valence connectivity index of order 0. (), ii. Valence connectivity index of order 1. ), iii. Valence connectivity index of order 2. (), iv. Shape index of order 1. (), v. Shape index of order 2. (), vi. Shape index of order 3. () The values of descriptors have been evaluated by solving the relevant equations given in theory. For QSAR prediction, the 3D modeling and geometry optimization of all the derivatives and evaluation of values of descriptors have been done with the help of CAChe Pro of Fujitsu software using the semiemperical PM3 Hamiltonian. The Project Leader program associated with CAChe Pro has been used for multi linear regression (MLR) analysis. The statistical parameters have been calculated by Smith’s Statistical Package (version 2.80). Results and DiscussionForty derivatives of testosterone given in table-1 have been considered for QSAR study where biological activity has been reported in terms of SHBG-ligand dissociation parameter (pKd). The values of six descriptors along with their biological activities are given in table-2. For the development of QSAR models multi linear regression (MLR) analysis has been performed using different combinations up to four descriptors. Compound no. - 31 and 32 are outlier in the MLR analysis. Three QSAR models with good predictive power have been found from single descriptors namely shape index of order 1, 2 and 3. The MLR equations for these three models are given below, SinglePA1 = -0.51247* + 16.343. r2 = 0.583013, rCV2 = 0.559699, Std. error = 0.1409, SEE = 0.7586, t-value = 7.0952, p-value=0, DOF = 0.5715. SinglePA2 = -1.71965* + 17.1998. r2 = 0.736069, rCV2 = 0.709896, Std. error = 0.0998, SEE = 0.6035, t-value = 10.0217, p-value=0, DOF = 0.7288. SinglePA3 = -4.61117* + 18.0597. r2 = 0.776952, rCV2 = 0.770113, Std. error =0.0893, SEE = 0.5548, t-value = 11.1088, p-value=0, DOF = 0.7708. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Parent Skeleton Structure of Testosterone Fig-1In the above regression equations, r2 is correlation coefficient, rCV2 is cross-validation coefficient, Std. Error is standard error, SEE is standard error of estimate and DOF is degrees of freedom. Shape indices appear important descriptors for this set of testosterone derivatives. All the above three models developed from single descriptor are good. Among these three models, QSAR model developed from shape index (of order 3) has the highest predictive power on the basis of value of correlation coefficient, cross-validation coefficient and statistical parameters. The trend of observed activity and predicted activity obtained from SinglePA3is shown in figure-2. On the other hand, valence connectivity index of order 0, 1 and 2 individually fail to give reliable QSAR model. The predicted activities SinglePA1, SinglePA2 and SinglePA3, obtained by substituting the values of descriptor shown in table-2 in the above regression equations, are listed in table-3. Various QSAR models, using descriptors in different combinations, have been developed but only top five models are reported. The best QSAR model (PA1) has been obtained by following regression equation, PA1 = -0.578992* + 2.8754* - 9.9885* + 0.54334* + 15.9834. r2 = 0.822955, rCV2 = 0.782252, Std. error = 0.0773, SEE = 0.4944, t-value = 12.9350, p-value=0, DOF = 0.8180. Figure-1 Parent Skeleton Structure of Testosterone Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 49 Descriptors used in this QSAR model are shape index (order 1), shape index (order 2), shape index (order 3) and valence connectivity index (order 0). Value of correlation coefficient, cross-validation coefficient and statistical parameters indicate that the predictive power of this QSAR model is very good. The trend of observed activity and predicted activity obtained from PA1 is shown in figure-3. The second best QSAR model (PA2) has been obtained by following regression equation, PA2 = -0.27818* - 4.86353* + 0.82815* - 0.532034* + 16.0292. r2 = 0.807879, rCV2 = 0.740665, Std. error = 0.0813, SEE = 0.5150, t-value = 12.3037, p-value=0, DOF = 0.8025. This regression equation has been developed by using descriptors shape index (order 1), shape index (order-3), valence connectivity index (order-0) and valence connectivity index (order-2). Value of correlation coefficient, cross-validation coefficient and statistical parameters indicate that this model has good predictive power. The trend of observed activity and predicted activity obtained from PA2 is shown in figure-4. The third best QSAR model (PA3) has been obtained by following regression equation, PA3 = -0.333292* - 3.87646* + 0.801008* - 0.685657* + 16.7645. r2 = 0.805092, rCV2 = 0.724064, Std. error = 0.0820, SEE = 0.5187, t-value = 12.1939, p-value=0, DOF = 0.7997. This regression equation has been developed by using descriptors shape index (order 1), shape index (order-3), valence connectivity index (order-0) and valence connectivity index (order-1). Value of correlation coefficient, cross-validation coefficient and statistical parameters indicate that this model has good predictive power. The trend of observed activity and predicted activity obtained from PA3 is shown in figure-5. The fourth best QSAR model (PA4) has been obtained by following regression equation, PA4 = 0.808584* - 7.82536* + 0.478292* - 0.392411* + 17.4375. 2 = 0.800686, rCV2 = 0.764064, Std. error = 0.0831, SEE = 0.5245, t-value = 12.0268, p-value=0, DOF = 0.7952.Descriptors used in this QSAR model are shape index (order 2), shape index (order 3), valence connectivity index (order 0) and valence connectivity index (order 2). Value of correlation coefficient, cross-validation coefficient and statistical parameters indicate that the predictive power of this QSAR model is good. The trend of observed activity and predicted activity obtained from PA4 is shown in figure-6. The fifth best QSAR model (PA5) has been obtained by following regression equation, PA5 = -6.50984* + 0.524319* + 0.723604* - 0.908735* + 16.1207. r2 = 0.800100, rCV2 = 0.725212, Std. error = 0.0833, SEE = 0.5253, t-value = 12.0042, p-value=0, DOF = 0.7946. This regression equation has been developed by using descriptors shape index (order 3), valence connectivity index (order-0), valence connectivity index (order-1) and valence connectivity index (order-2). Value of correlation coefficient, cross-validation coefficient and statistical parameters indicate that this model has good predictive power. The trend of observed activity and predicted activity obtained from PA5 is shown in figure-7. The predicted activities (PA1-PA5) obtained from above top five QSAR models are listed in table-4. From the values of correlation coefficient (r), cross-validation coefficient (rCV) and statistical parameters for the above five QSAR models, it is clear that the predictive power of all models is high and can be used to find out the SHBG activity of any derivative of testosterone. The correlation summary of top five QSAR models is presented in table-5. Conclusion It is clear from the above study that, the best combination of descriptors is shape index (order 1), shape index (order 2), shape index (order 3) and valence connectivity index (order 0) for QSAR study of testosterone derivatives as SHBG ligand. Three reliable QSAR models have been obtained from single descriptor namely shape index (order 1), shape index (order 2) and shape index (order 3). Whereas, valence connectivity indices (of order 0, 1 and 2) individually fail to give any reliable QSAR model. Therefore, shape indices of order 1, 2 and 3 appear good descriptor for QSAR study of testosterone derivatives. Among these three descriptors, shape index (order 3) is the best because the value of correlation coefficient is 0.776952 and value of cross-validation coefficient is 0.770113 for the QSAR model obtained from this descriptor. Also, shape index (order 3) is present in combination with other descriptors in all best five QSAR models. References 1.Wilson J.D., Griffin J.E. and George F.W., Sexual differentiation: early hormone synthesis and action, Biol. Reprod.22, 9-17 (1980) 2.Westphal U., Steroid-Protein Interactions II, Monographs on Endocrin27, 603 (1986)3.Dunn J.F., Nisula B.C. and Rodbard D., Transport of steroid hormones: binding of 21 endogenous steroids to both testosterone-binding globulin and corticosteroid-binding globulin in human plasma, J. Clin. Endocrinol. Metab.53, 58-68 (1981) Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 50 Table-1 Forty derivatives of testosterone with their biological activity in terms of SHBG-ligand dissociation parameter (pKd) C. No. Name of Compound pKd 1 5--Dihydrotestosterone 9.74 2 1--methyl-Dihydrotestosterone 9.60 3 5-Androstene-19-nor-3,17-diol 9.54 4 -methyl-Dihydrotestosterone 9.38 5 2-Iodo Estradiol 9.32 6 4-methyl-Testosterone 9.31 7 Testosterone 9.20 8 -Androstene-3, 17-diol 9.17 9 -Androstane-3, 17-diol 9.11 10 2-Methoxy Estradiol 9.08 11 -methyl-14-dehydro-19-Nor-testosterone 9.07 12 -17-dimethyl-Dihydrotestosterone 9.05 13 -methyl-5-Androstane-3, 17-diol 9.00 14 4-Androstene-3, 17-diol 9.00 15 Estradiol 8.83 16 17-methyl-Dihydrotestosterone 8.81 17 , 17-dimethyl-Testosterone 8.76 18 6-dehydro-Estradiol 8.76 19 -methyl-Testosterone 8.71 20 7-dehydro-Estradiol 8.62 21 17-methyl-1-Dihydrotestosterone 8.57 22 , 17-dimethyl-5-Androstane-3, 17-diol 8.46 23 -methyl-19-nor-Dihydrotestosterone 8.46 24 17-methyl-Testosterone 8.43 25 -methyl-Testosterone 8.36 26 19-Nor-Dihydrotestosterone 8.36 27 -methyl-1-dehydroTestosterone 8.36 28 17-Deoxoestrone 8.30 29 Estrone 8.18 30 16-hydroxyTestosterone 7.92 31 Dehydroepiandrosterone 7.84 32 Androstenedione 7.46 33 Deoxycortisol 7.44 34 Deoxycorticosterone 7.38 35 Pregnenolone 7.15 36 Cortisone 6.43 37 Corticosterone 6.34 38 Cortisol 6.20 39 Aldosterone 5.32 40 17-Aminopropyl-17-hydroxy-5-Androstan-3-one 4.96 Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 51 Table-2 Values of descriptors and observed activities of forty derivatives of testosterone C. No. 0  1  2 pKd 1 14.583 4.747 1.926 13.606 9.147 9.037 9.74 2 15.523 4.997 1.977 14.476 9.568 9.503 9.60 3 13.648 4.750 1.961 12.593 8.644 8.128 9.54 4 15.523 4.997 2.042 14.476 9.558 9.587 9.38 5 14.583 5.000 2.066 14.637 9.277 8.718 9.32 6 15.523 4.997 1.977 14.322 9.292 8.997 9.31 7 14.583 4.747 1.926 13.399 8.870 8.607 9.20 8 14.583 4.747 1.926 13.463 9.004 8.817 9.17 9 14.583 4.747 1.926 13.722 9.311 9.223 9.11 10 15.523 5.523 2.259 13.509 8.622 7.776 9.08 11 14.583 5.000 2.066 13.140 8.624 8.069 9.07 12 16.467 5.011 2.083 15.399 9.930 10.218 9.05 13 15.523 4.997 2.042 14.593 9.722 9.772 9.00 14 14.583 4.747 1.926 13.515 9.023 8.769 9.00 15 13.648 4.750 1.961 12.179 8.093 7.436 8.83 16 15.523 4.762 1.977 14.529 9.519 9.668 8.81 17 16.467 5.011 2.083 15.192 9.652 9.794 8.76 18 13.648 4.750 1.961 11.919 7.786 7.100 8.76 19 15.523 4.997 2.042 14.269 9.280 9.163 8.71 20 13.648 4.750 1.961 11.971 7.792 7.034 8.62 21 15.523 4.762 1.977 14.062 8.945 8.872 8.57 22 16.467 5.011 2.083 15.515 10.094 10.403 8.46 23 14.583 5.000 2.066 13.554 9.202 8.904 8.46 24 15.523 4.762 1.977 14.322 9.242 9.238 8.43 25 15.523 4.997 2.042 14.269 9.290 9.135 8.36 26 13.648 4.750 1.961 12.684 8.791 8.355 8.36 27 15.523 4.997 2.042 14.010 8.984 8.797 8.36 28 12.719 4.500 1.919 11.861 7.991 7.384 8.30 29 13.648 4.750 1.961 12.062 7.945 7.217 8.18 30 15.523 4.997 2.042 13.717 8.961 8.761 7.92 31 14.583 4.747 1.926 13.399 8.859 8.574 7.84 32 14.583 4.747 1.926 13.283 8.722 8.387 7.46 33 18.367 6.000 2.371 15.384 9.866 9.405 7.44 34 17.416 6.021 2.481 15.014 9.774 9.268 7.38 35 16.467 5.500 2.289 14.976 9.741 9.441 7.15 36 19.322 6.250 2.486 15.585 9.804 9.381 6.43 37 18.367 6.270 2.588 15.332 9.865 9.420 6.34 38 19.322 6.250 2.486 15.702 9.957 9.557 6.20 39 19.322 6.805 2.704 15.317 9.890 9.144 5.32 40 18.367 6.270 2.588 16.227 10.781 10.318 4.96 Where, 1 = Shape index (order 1), 2 = Shape index (order 2), 3 = Shape index (order 3), = Valence connectivity index (order 0), = Valence connectivity index (order 1), = Valence connectivity index (order 2). Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 52 Table-3 Predicted activities SinglePA1 to SinglePA3 of the testosterone derivatives C. No. SinglePA1SinglePA2SinglePA3 1 8.870 9.037 9.179 2 8.388 8.607 8.943 3 9.349 9.031 9.017 4 8.388 8.607 8.644 5 8.870 8.602 8.533 6 8.388 8.607 8.943 7 8.870 9.037 9.179 8 8.870 9.037 9.179 9 8.870 9.037 9.179 10 8.388 7.702 7.643 11 8.870 8.602 8.533 12 7.904 8.583 8.455 13 8.388 8.607 8.644 14 8.870 9.037 9.179 15 9.349 9.031 9.017 16 8.388 9.011 8.943 17 7.904 8.583 8.455 18 9.349 9.031 9.017 19 8.388 8.607 8.644 20 9.349 9.031 9.017 21 8.388 9.011 8.943 22 7.904 8.583 8.455 23 8.870 8.602 8.533 24 8.388 9.011 8.943 25 8.388 8.607 8.644 26 9.349 9.031 9.017 27 8.388 8.607 8.644 28 9.825 9.461 9.211 29 9.349 9.031 9.017 30 8.388 8.607 8.644 33 6.930 6.882 7.127 34 7.418 6.846 6.619 35 7.904 7.742 7.505 36 6.441 6.452 6.596 37 6.930 6.418 6.126 38 6.441 6.452 6.596 39 6.441 5.498 5.591 40 6.930 6.418 6.126 Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 53 Table-4 Predicted activities PA1 to PA5 of the testosterone derivatives C. No. PA1 PA2 PA3 PA4 PA5 1 9.344 9.065 9.065 9.166 9.123 2 9.482 9.028 8.962 9.202 9.129 3 8.994 8.800 8.774 8.766 8.826 4 8.833 8.667 8.717 8.660 8.622 5 9.234 9.408 9.259 8.893 9.136 6 9.399 9.170 9.028 9.327 9.308 7 9.232 9.122 9.089 9.235 9.205 8 9.267 9.064 9.048 9.184 9.145 9 9.407 9.062 9.045 9.148 9.134 10 7.652 7.775 7.743 7.636 7.671 11 8.420 8.513 8.507 8.432 8.469 12 8.419 8.634 8.728 8.545 8.535 13 8.897 8.666 8.698 8.644 8.634 14 9.295 9.132 9.077 9.227 9.229 15 8.769 8.825 8.820 8.840 8.839 16 8.835 8.984 9.038 8.972 8.971 17 8.306 8.688 8.752 8.612 8.610 18 8.628 8.788 8.823 8.847 8.786 19 8.720 8.721 8.742 8.728 8.697 20 8.656 8.867 8.860 8.898 8.878 21 8.582 9.021 9.058 9.061 9.034 22 8.482 8.632 8.708 8.528 8.546 23 8.645 8.412 8.443 8.302 8.345 24 8.723 9.042 9.062 9.042 9.053 25 8.720 8.736 8.735 8.739 8.730 26 9.044 8.754 8.746 8.721 8.774 27 8.580 8.702 8.737 8.747 8.680 28 8.835 9.052 9.108 8.835 8.920 29 8.706 8.845 8.828 8.870 8.870 30 8.421 8.478 8.518 8.621 8.543 33 7.277 7.125 7.010 7.402 7.344 34 6.589 6.621 6.667 6.435 6.492 35 7.537 7.695 7.720 7.431 7.541 36 6.404 6.479 6.449 6.810 6.678 37 5.858 6.018 6.128 5.892 5.890 38 6.467 6.482 6.438 6.797 6.690 39 5.677 5.323 5.331 5.518 5.396 40 6.344 6.282 6.217 5.968 6.206 Table-5 Correlation Summary of best five QSAR models S. No. QSAR Model rCV Std. Error SEE t -value p- value DOF VC 1 PA1 0.822955 0.782252 0.0773 0.4944 12.9350 0 0.8180 4 2 PA2 0.807879 0.740665 0.0813 0.5150 12.3037 0 0.8025 4 3 PA3 0.805092 0.724064 0.0820 0.5187 12.1939 0 0.7997 4 4 PA4 0.800686 0.764064 0.0831 0.5245 12.0268 0 0.7952 4 5 PA5 0.800100 0.725212 0.0833 0.5253 12.0042 0 0.7946 4 Where; r=Correlation coefficient, rCV=Cross-validation coefficient, Std. Error = Standard error, SEE=Standard error of estimate, DOF=Degrees of freedom, VC=Variable count Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 54 Figure-2 Trend of observed activity (pKd) and predicted activity (obtained from SinglePA3) of the testosterone derivatives Figure-3 Trend of observed activity (pKd) and predicted activity (obtained from PA1) of the testosterone derivatives Figure-4 Trend of observed activity (pKd) and predicted activity (obtained from PA2) of the testosterone derivatives Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 55 Figure-5 Trend of observed activity (pKd) and predicted activity (obtained from PA3) of the testosterone derivatives Figure-6 Trend of observed activity (pKd) and predicted activity (obtained from PA4) of the testosterone derivatives Figure-7 Trend of observed activity (pKd) and predicted activity (obtained from PA5) of the testosterone derivatives Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 3(5), 47-56, May (2013) Res. J. Chem. Sci. International Science Congress Association 56 4.Hammond G.L., Access of reproductive steroids to target tissues, Obstet. Gynecol. Clin. North. Am.,29, 411-423 (2002) 5.Hammond G.L. and Bocchinfuso W.P., Sex hormone-binding globulin/androgen-binding protein: steroid-binding and dimerization domains, J. Steroid. Biochem. Mol. Biol.,53, 543-552 (1995)6.Hogeveen K.N., Cousin P., Pugeat M., Dewailly D., Soudan B. and Hammond G. L., Human sex hormone-binding globulin variants associated with hyperandrogenism and ovarian dysfunction, J. Clin. Invest.,109, 973-981 (2002) 7.Van Pottelbergh I., Goemaere S., Zmierczak H. and Kaufman J. M., Perturbed sex steroid status in men with idiopathic osteoporosis and their sons, J. Clin. Endocrinol. Metab.,89, 4949-4953 (2004) 8.Rapuri P.B., Gallagher J.C. and Haynatzki G., Endogenous levels of serum estradiol and sex hormone binding globulin determine bone mineral density, bone remodeling, the rate of bone loss, and response to treatment with estrogen in elderly women, J. Clin. Endocrinol. Metab., 89, 4954-4962 (2004) 9.Cherkasov A., Shi Z., Li Y., Jones S.J., Fallahi M. and Hammond G.L., Inductive charges on atoms in proteins: comparative docking with the extended steroid benchmark set and discovery of a novel SHBG ligand, J. Chem. Inf. Model., 45, 1842- 1853 (2005) 10.Cherkasov A., Inductive Descriptors. 10 Successful Years in QSAR, Curt. Comp-Aided Drug Design,, 21-42 (2005) 11.Singh Rajeev, Kumar D., Singh Bhoop, Singh V.K. and Sharma Ranjana, Molecular structure, vibrational spectroscopic and HOMO, LUMO studies of S-2-picolylN-(2-acetylpyrrole) dithiocarbazate Schiff base by Quantum Chemical investigations, Research Journal of Chemical Sciences, 3(2), 79-84, (2013) 12.Gupta Y.K., Agarwal S.C., Madnawat S.P. and Ram Narain, Synthesis, Characterization and Antimicrobial Studies of Some Transition Metal Complexes of Schiff Bases, Research Journal of Chemical Sciences, 2(4), 68-71, (2012) 13.Buttrus H. Nabeel and Saeed T. Farah, Synthesis and Structural Studies on Some Transition metal complexes of Bis-(benzimidazole-2-thio) ethane, propane and butane ligands, Research Journal of Chemical Sciences, 2(6), 43-49, (2012) 14.Gupta Manish and Sharma Vimukta, Targeted drug delivery system: A Review, Research Journal of Chemical Sciences, 1 (2), (2011) 15.Glossman-Mitnik D., Computational Molecular Nanoscience: A Study of the Molecular Structure and Properties of a RAFT Polymerization Agent, Research Journal of Chemical Sciences,1(9), 6-10, (2011) 16.Kier L. B., Shape indexes for orders one and three from molecular graphs, Quant. Struct.-Act. Relat.,, 1-7 (1986) 17.Kier L.B., Indexes of molecular shape from chemical graphs, Med. Res. Rev., 417-440 (1987) 18.Randic M., On characterization of molecular branching, J. Am. Chem. Soc., 97, 6609-6615 (1975) 19.Kier L.B. and Hall L.H., Molecular Connectivity in Chemistry and Drug Research, Wiley, New York. (1986) 20.Singh P.P., Srivastava H.K. and Pasha F.A., DFT-based QSAR study of testosterone and its derivatives, Bioorg. Med. Chem., 12, 171 (2004) 21.Srivastava H.K., Pasha F.A. and Singh P.P., Atomic softness-based QSAR study of testosterone, Int. J. Quant. Chem. 103, 237-245 (2005) 22.Srivastava H.K., Pasha F.A., Mishra S.K. and Singh P.P., Novel applications of atomic softness and QSAR study of testosterone derivatives, Med. Chem. Res.,18, 455-466 (2009) 23.Balaban A.T., Graph theory and theoretical chemistry, Jour. of. Mol. Struc: THEOCHEM,120, 117-142 (1985) 24.Petitjean M., Applications of the Radius-Diameter Diagram to the Classification of Topological and Geometrical Shapes of Chemical Compounds, J. Chem. Inf. Comput. Sci.,32, 331-337 (1992)