Intelligent consensus prediction for addressing ecotoxicological effects of diverse pesticides on California quail

Abstract

Birds occupy a major portion of the ecology and are considered a valuable species. In this modern era, the application of pesticides has increased and caused very severe harmful consequences to various non-target species, including birds. Many researchers have reported that the number of endangered bird species has been increasing day by day owing to the harmful effects of chemical pesticides. Restoration and protection of various endangered avian species from exposure to potentially hazardous pesticides pose a challenge from the standpoint of ecosystem safety evaluation. In the current study, partial least squares (PLS)-based quantitative structure toxicity relationship (QSTR) models were generated to enable the prediction of pesticide toxicity towards California quail. “Intelligent consensus prediction” (ICP) was also performed to increase the external predictability of the constructed models. A pesticide database (Pesticide properties DataBase) consisting of 1694 pesticides was screened by employing the developed PLS-based QSTR models. From the developed models, we found that the presence of a phosphate moiety, high percentage of carbon, and electronegativity are responsible for increasing the toxicity. In contrast, the presence of a greater number of rotatable bonds, multiple bonds, aromatic proportion, and molecular polarity diminish the toxicity. The data derived from the generated chemometric models might be beneficial for the various new and untested chemical pesticides. These models may offer guidance to future researchers to fabricate novel and eco-friendly pesticides and data-gap filling.