Flavonoids identified in Citrus fruits cover over 60 types of flavonoids, according to the subclasses of flavone derivatives. Among the subclasses of flavones, a series of naturally occurring or synthetic analogous of polymethoxylated flavones have been shown in numerous studies to exert strong antiproliferative actions against a range of number of human cancer cell lines [1-2]. Quantitative structure-activity relationship (QSAR) analysis has been carried out to a data set of 17 polymethyloxylated flavones in citrus against six human cancer cell lines including lung, prostate, colon, melanoma, and estrogen receptor positive (ER+) and estrogen receptor negative (ER-) breast cancer to correlate and predict the antiproliferative activities. Modeling of the antiproliferative activities of these compounds as a function of the theoretically derived descriptors was established by multiple linear regression analysis (MLRA). The genetic algorithm (GA) was used for the selection of the variables that resulted in the best-fitted models. The obtained equations consist of molecular descriptors calculated from the characteristics of the molecular structures with use of DRAGON software [3] and quantum-chemical methods. A number of molecular descriptors were obtained from the stepwise selection procedure. The performances of the QSAR model for each type of human cancer cell line were investigated by obtaining the R2 values for basic model. The Q2 (cross validation R2) values and scrambling/randomization experiments also confirm the statistical significance of our models.
