Szeged-like topological descriptors and COM-polynomials for graphs of some Alzheimer's agents

In this study, we define Szeged-like topological coindices in which formulas are constructed by a complement of a given graph. We introduce two new co-polynomials of two variables, denoted by SMP(x,y) and SMPe(x,y) briefly, to compute modifications and edge versions of these molecular descriptors. In order to the perform the effect of the topological descriptors, firstly we obtain the graph structures of five novel diamide derivatives which are considered potential agents for Alzheimer's disease. After completing the computation process of each index from some modifications of the polynomials mentioned above, we correlate the coindices values with the pMIC values of the compounds against three Gram-negative bacteria E. coli, P. aeruginosa, K. pneumoniae, and three Gram-positive bacteria E. faecalis, B. cereus, S. aeurus to predict the bioactivity properties. As a result of the QSAR analysis, we get good correlations among the bioactivity data and topologic values which shows the effectiveness of our newly created coindices. Also we give mathematical equations giving the best approach to predict the relation between the topological coindices and bioactivity properties obtained by exponential regression analysis. [GRAPHICS] .