In Silico Treatment for Prediction of New Effective Pesticides Based on Acetylcholinesterase Inhibition

Abstract

The presented paper introduces in silico treatment for 34 commercially available pesticides in order to predict new compounds that may act as pesticides based on acetylcholinesterase inhibition. Theoretical treatments using density functional theory, (DFT) and quantitative structure–activity relationship, (QSAR) suggested two statistically significant models for the median lethal dose of Mus musculus, (MLD50) and Homo sapiens, (HLD50). The MLD50 model consisted of only two descriptors including the number of hydrogen bond donor (NHBD) and ovality (OVA) in addition to the correction term (ZOM) with the value of 0.905 for both square correlation coefficient (r²) and cross-validation (q²). The model of HLD50 was only depending on three descriptors including NHBD, OVA, and Wiener index (WI) in addition to ZOM with r² and q² equal to 0.853 and 0.777. The proposed equations showed a physical meaning which could help in understanding the factors affecting the action of pesticides. A special and unique feature was introduced in this study by exploiting both developed models of MLD50 and HLD50 in predicting new more effective pesticides with less toxicity to humans. This new approach was established by the correlation between those models which in turn seven new specific compounds that might be usable as new pesticides were introduced.