Predicting the neurotoxic activity of diverse bio/chemical compounds on Acetyl cholinesterase enzyme using deep learning-based QSAR model

A deep learning model that utilizes neural network architectures to predict bioactivity against Acetylcholinesterase (AChE) enzyme was applied. The model effectively captures complex relationships within high-dimensional data by integrating neural network architecture training and optimization techniques. The obtained results indicate that the deep learning model achieves high accuracy, sensitivity, and precision in the predictions. The enhanced performance with a low loss score can be attributed to the ability of the model to leverage the unique capabilities of deep learning computing, enabling it to explore a broader solution space. This work demonstrates that the use of computational modeling by means of deep learning provides a novel, rapid, accurate, and cost-effective method over traditional neurotoxic test alternatives in prediction, which can be used as a novel candidate for the neurotoxic effects of biochemical compounds on AChE, which is a potential biomarker used in neurotoxic research. This study highlights the potential of neural networks as a modeling approach for quantitative structure-activity relationship (QSAR) modeling, paving the way for advancements in drug discovery and toxicological research.