QSAR Research of Triazolopyrimidine Herbcides by Electronegativity-distance Vector and Artificial Neural Network

To study the quantitative structure-activity relationship(QSAR) of the inhibited activity(pI₅₀) for aceto lactate synthase(ALS) inhibitor,molecular electronegativity-distance vector(M_k) was used to describe the molecular structure of 31 triazolopyrimidine herbcides in this paper. The sixparameter(m²、M₁₀、M₁₄、M₁₅、M₆₇、M₈₅) QSAR model of pI₅₀ for 31 the triazolopyrimidine compounds was constructed by leaps-and-bounds regression(LBR). The traditional correlation coefficient(R) and the cross-validation correlation coefficient(R_CV) were 0.911 and 0.887,respectively. The six structural parameters were used as the input neurons of artificial neural network,and a 6∶ 3 ∶ 1 network architecture was employed. A satisfied model could be constructed with the back-propagation algorithm,the correlation coefficient R was 0. 988. The result demonstrates that the dominant influencing factors of inhibited activity are the molecular structure fragments:-CH₃、-CH₂-、> C-、-O-、> N-及-X(-F,-Cl),and there is a good non-linear relationship between the parameter m²、M₁₀、M₁₄、M₁₅、M₆₇、M₈₅) QSAR model of pI₅₀ and pI₅₀ of ALS. The model can provide some theoretical insights into the design of this series of ALS inhibitor with higher inhibited activity.