Qualitative Structure Activity Relationship - Report Example

A qualitative structure activity relationship (QSAR) studies A qualitative structure activity relationship (QSAR) studies have been performed on numerous structured experiments including that aimed at determining the relationship between electronic parameters and topochemical and two measured activities. These steroid hormones studies on steroid hormones shows that homones have certain degree of binding affinity to the corticosteroid binding globulin (CBG) and gestagenic receptors. The data reduction was performed based on principle component analysis (PCA). The loaded values are expected to give descriptive information about the descriptors that will be interpreted in terms of significant biological correlation relationships. Notably, the experimental model applied in this study revealed that the main features describing the binding activities of the experimental compounds were partially charge-based receptors. This was an automatic measure of molecular electronic property. The same property can be used in an auto correlated that can be used to weigh substitute descriptors’ position. Additionally, the linear regression equation and its application can be used in predicting biological activity in the study. It is worth noting that the comparison of the result with the literature data indicates that the proposed quality of the model compared to the provided model compares well with the sophisticated 2D based method. Introduction The quantitative structured activity relationship, the QSAR techniques are currently suitable and indispensable in all research aspects especially in relation to biological molecular properties interpretation. Obviously, the chemical, biological, and or physical properties of a compound depend on the three dimensional (3D) molecular and atomic arrangement; however, the same properties can be examined in a two dimensional model. Notably, the capacity to produce qualitative and quantitative correlation between 2D molecular structures and their biological activities is fundamental in deciding what synthetic ways of bioactive chemicals. Numerous research works involving the application and use of QSAR have been applied in biological properties and structural determination of biological elements; for instance, the use of the Hansch molecule description. The QSARs analysis of the 2, 4- diamino – 5- (substituted benzyl) pyrimidines and 2,4 – diamino – 6 – dimethyl – 5 – phenyldihydrotriazines have suggested that neural network application have reliable and effective results than normal or traditional application of the regression methods. In supplementation with the regression processes, other techniques of determination of the activity of the biological molecules have since been introduced. This includes the use of inductive logic programming (ILP) that has since used to model the QSAR particularly in the trimethoprim analogues binding from E. coli to DHFR. This technique applies mainly the physiochemical attributes that are usually assigned to substituent heuristically towards a general approach. This application or process is usually applied to the design problem of drugs. However, it should be noted that this method is not significantly better than the regression or the traditional QSAR method; however, it is noted to produce rules that are likely to provide insight into the stereochemistry of the biological molecules under determination. In most cases, such stereochemistry does help in determining the structure and to some extent, depending on the analysis, chemical properties of a molecule. It should be noted that more evidence are still required to assess these newly applied methods properly to ascertain the comparative trials and the analysis in numerous and different ways. The biological steroids activities usually vary considerably depending on seemingly small structural changes within or about a molecule. The fundamental molecular family usually represents very challenging characteristics for any method of prediction. This dependency is usually due to relatively cyclopentanoperhydrophenanthrene skeleton low flexibility2. Therefore, numerous perfect and excellent QSAR models that are based on 2D properties including topological descriptors have been determined to have quality compared to the results provided by sophisticated and complex 3D based methods. Nonetheless, this paper aims at determining the Application of QSAR activity techniques to Identify molecular properties of substituent at position X as an important for inhibitory activity and Derive quantitative models that can be used for rational design of HSV-2 inhibitors. Material and Method A data set of steroids with binding globulin (CBG) affinity was used in this analysis. The selected set was repeatedly served as the experimental benchmark used in evaluating the QSAR method performance. The structured applied in this experiment are carefully analyzed against any devoid or errors that may affect the results. The project was mainly qualitative; thus, it used light substituent including hydroxyls and oxygen at the active site X. It should be noted that the active site usually increases the CBG’s activity and the availability of the bulk chain including COCH2OH increases or enhances the activity. The QSAR methods in this study is to be conducted using fivefold cross validation were the data set will be divided randomly into equal sets. Each of the data sets will be studied in correspondence to training sets. From the literature data, it is expected the cross validation will appear once in only a single data test sets. Notably, the divisions will maintain consistency as the earlier QSAR analytical methods1. The analysis will deploy mainly the linear regression using the Hansch parameters including πs, MRs, and sigma. These elements were designated per molecule within the training sets. The benchmark in this study shall be achieved through a stepwise linear regression that will be proved on the square of the underlying variables. However, the regression will be obtained from full automation from a step commands with a Minitab. It should be noted the predictor variables would be included in the regression process in accordance to the criterion of the maximum F-Statistics. Results SUMMARY OUTPUT Regression Statistics Multiple R 0.912621 R Square 0.832877 Adjusted R Square 0.795739 Standard Error 0.237386 Observations 12 Table 1: results for statistical regression ANOVA   df SS MS F Significance F Regression 2 2.527533 1.263766 22.42633 0.000319 Residual 9 0.507167 0.056352 Total 11 3.0347       Table 2: ANOVA results   Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept 5.362208 0.126278 42.46344 1.11E-11 5.076547 5.647869 SC(PI) 0.410125 0.089397 4.587662 0.001313 0.207894 0.612355 SC(MW) 0.000427 0.000369 1.157684 0.276792 -0.00041 0.001262 Table 3: coefficient correlation results Many applications have since been used to determine descriptors most of which have indicated that the QSAR studies are of great importance in the biological structure activity relationship. The computation of these descriptors are usually done through numerous software products; however, this research deployed the use of excel software that provided the above results5. The test group of these descriptors was based on the two – dimensional autocorrelation function that were applied to a molecular graph. The descriptors were found to express a correlation between statistical weight and numerical values that were determined to vary across various atomic properties. It should be noted that the values obtained from this study directly reflect to high degree the effects of the Sanderson electro-negativity are depicted from the weighting coefficients. Additionally, some good amounts of charges are predicated to concentrate inside the molecule. Therefore, from the geometric descript, the descriptors indicated the molecule sizes that in most cases are driven from the atomic nuclei three-dimensional coordinates. Some other factors that determine the geometry of such descriptors include the atomic radii and or atomic mass of atoms with the molecule. Discussion There is usually intensive complexity especially in the interaction of molecular receptors and their potential molecule inhibitors; thus, the simple linear regression using the 2D model may not actually be accurate in predicting or determining the actual structure among other properties. However, the properties of the statistical techniques applied in this case are usually vital in eliminating uncorrelated variables3. Additionally, loading factors usually plays a significant role in evaluating the relevant receptor within the system especially those that are considered to be contributing highly to the variance of the data. High loading values usually indicate the principle component (PC)3 as aligned to the direction close to the response of the original receptor. Each principle component is usually analyzed to help in determining how the descriptor contributes to the PC itself. Loading plots an or table as above usually provide vital information on what descriptors can provide unique information and repeat the information on different element or character under study. For accurate results, the QSAR model is to be validated against an external predicting set. The model calculation predicts the actual partial charge within the analytical molecule and the same should be determined prior to applying the training set. In this case, it is vital to assume that biological activity in the analysis is unknown. Evaluating from the graph below, it is apparent that the analysis was carried out with high degree of accuracy. The coefficient values 5.076547 and 5.647869 for lower and upper values respectively shows that meta-substituted phenylguanine activity in analogous for the activities of HSV 1 and HSV 2 thymidine kinase inhibitors6. However, the functionally of HSV 1 and HSV 2 thymidine kinase inhibitors differ with the element or compound found at position X. notable, the same element at this position is what make the coefficient values of the study to differ significantly. Conclusion To explain the contribution of each position of substituent to the binding affinity of a receptor, it would be vital to generate initially a simple electronic descriptor defined by atomic partial charge in correlation with other study properties. The QSAR model applied in this study depicts that global descriptor is nearly among the most significant factors in predicting the activities of biological molecular compounds. In this case, it can be applied in determining the substituent positions. Therefore, calculating the atoms positions as per this study, accounts for 95% of the variance, this result confirms that the molecular properties of the chemical groups at position X that are important in determining the inhibitory activity of the analogues. Hence, the quantitative models that can be used for the rational design of HSV thymidine kinase inhibitors since the applied model has validated the external prediction test. Finally, it is worth noting that the simple 2D models developed in this study are equally comparable to the sophisticated 3D based methods; however, the latter requires more conceptualized resources. References 1. Top of Form 1. Puzyn, Tomasz, Jerzy Leszczynski, and Mark T. D. Cronin. Recent Advances in Qsar Studies: Methods and Applications. Dordrecht: Springer, 2010. 2. Top of Form 2. Merz, Kenneth M, Dagmar Ringe, and Charles H. Reynolds. Drug Design: Structure- and Ligand-Based Approaches. Cambridge [U.K.: Cambridge University Press, 2010. Top of Form 3. Patrick, Graham L, and John Spencer. An Introduction to Medicinal Chemistry. New York: Oxford University Press, 2009. 3. Top of Form 4. Kubinyi, Hugo. Qsar: Hansch Analysis and Related Approaches. Weinheim: VCH, 1993. 4. Top of Form 5. Liang, Xiaotian, and Wei-Shuo Fang. Medicinal Chemistry of Bioactive Natural Products. Hoboken, N.J: Wiley-Interscience, 2006. Top of Form 5. Silverman, Richard B. The Organic Chemistry of Drug Design and Drug Action. Burlington: Elsevier Science, 2004. Bottom of Form Bottom of Form Bottom of Form Bottom of Form Bottom of Form Bottom of Form Read More