Novel QSPR modeling of stability constants of metal-thiosemicarbazone complexes by hybrid multivariate technique: GA-MLR, GA-SVR and GA-ANN

Abstract The quantitative structural property relationship (QSPR) models of the log β 11 stability constants of M:L complexes of the structurally diverse thiosemicarbazones and several metal ions (M = Ag + , Cd 2+ , Co 2+ , Cu 2+ , Fe 3+ , Mn 2+ , Cr 3+ , La 3+ , Mg 2+ , Mo 6+ , Nd 3+ , Ni 2+ , Pb 2+ , Zn 2+ , Pr 3+ , Dy 3+ , Gd 3+ , Ho 3+ , Sm 3+ , Tb 3+ , V 5+ ) in aqueous solution have been constructed by combining the genetic algorithm with multivariate linear regression (QSPR GA-MLR ), support vector regression (QSPR GA-SVR ) and artificial neural network (QSPR GA-ANN ). The multi-levels optimization for grid search technique is used to find the best QSPR GA-SVR model with the optimized parameters capacity C = 1.0, Gamma, γ = 1.0 and Epsilon, e  = 0.1. The quality of the QSPR models presented in statistical values as training R 2 in range 0.9148–0.9815, validation Q 2 in range 0.7168–0.9669 and MSE values in range 0.2742–2.4906. The new two thiosemicarbazone reagents were designed and synthesized based on the lead thiosemicarbazone reagents. The log β 11 values of new complexes Cu 2+ L, Ni 2+ L, Cd 2+ L and Zn 2+ L derived from the QSPR GA-SVR and QSPR GA-ANN model turn out to be in a good agreement with experimental data.