A quantitative structure activity relationship (QSAR) study was developed in order to model the toxicity of dioxin, specifically toxicity of polyhalogenate /polychlorinated dibenzo-p-dioxins (25PHCDs/ PCDDs).The QSAR model was constructed, using Genetic Function Algorithm (GFA).The Quantum chemical descriptors were computed by density functional theory (DFT) at B3LYP/ 6-31G*. Model-1 with highest statistical significance was selected and it has squared correlation coefficient (R2) =0.971, Cross validated correlation coefficient (Q2) = 0.961 and external prediction ability R2pred = 0.885. The accuracy of the developed model was evaluated through cross-validation, an external test set, Y-randomization and applicability domain techniques. The result of the present study is expected to be useful to predict and identify other toxic compound or to synthesis non-toxic dioxins.
