Flavonoid-metal complexes have anticancer activities. However, the quantitative structure-activity relationship (QSAR) of flavonoid-metal complexes and their anticancer activities has not been known so far. Based on the 14 structures of flavonoid-metal complexes and their anticancer activities for HepG2 from the references, we optimised their structures using the density functional theory (DFT) method, and subsequently calculated 19 quantum chemical descriptors, such as dipole, charge, and energy. Then, we chose several quantum chemical descriptors that are very important for IC 50 which represents the anticancer activities of flavornoid-metal complexes for HepG2 through the stepwise linear regression method. Meanwhile, we obtained six new variables through the principal component analysis. Finally, we built QSAR models based on those important quantum chemical descriptors, six new variables as independent variables, and IC 50 as a dependent variable using an artificial neural network (ANN). At last, we validated the models using the experimental data from the references. The results show that models presented in this paper are accurate and predictive.
2014) QSAR study of flavonoid-metal complexes scavenging O 2Flavonoid-metal complexes have antioxidant activities. However, the quantitative structure-activity relationship (QSAR) of flavonoid-metal complexes and their antioxidant activities is not known. On the basis of 31 structures of flavonoid-metal complexes and their antioxidant activities scavenging O 2 ·− , we optimized their structures using the density functional theory method, and subsequently calculated 21 quantum chemistry descriptors such as dipole, charge, and energy. Then, we chose several quantum chemistry descriptors that are very important to the IC 50 of the antioxidant activities of flavonoid-metal complexes for scavenging O 2 ·− through stepwise linear regression. We obtained six new variables through the principal component analysis. Finally, we built QSAR models based on those important quantum chemistry descriptors, the six new variables as independent variables, and the IC 50 as the dependent variable using an artificial neural network. We validated the model using experimental data in references. These results show that the model in this article is reliable and predictive.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.