Quantitative Structure–Activity Relationships (QSARs) Study for KCNQ Genes (Kv7) and Drug Discovery

Abstract

Neuronal Kv7 (KCNQ) genes are important group of voltage-gated K⁺ channels, popularly known as the M-current or M-channel. These genes play a pivotal role in regulating neuronal excitability and access the risk associated with these neuronal excitability with reference to hypertension, diabetes and obesity. They have vital physiological functions in the nervous system, heart and inner ear. Failure in the functioning of Kv7 channels leads to severe genetic disorders. The drugs targeting Kv7 channels can be used in the treatment of neuronal channelopathies like chronic and neuropathic pain, deafness and mental illness. Owing to the prime importance of the KCNQ channels, novel quantitative structure–activity relationships (QSARs) models were built based on key descriptors using multiple linear regression (MLR) technique. A diverse set of 60 organic molecules were used to build the models which showed good experimental interaction with KCNQ. The models were validated by the external validation and standardization approach. The feed-forward neural network (FFNN) has also been used to establish the accuracy and efficiency of the models. When the FFNN is applied to the descriptors attained from the MLR, it exhibited a correlation coefficient of 0.839. A high correlation has been observed between predicted and experimental pEC50(M), thereby confirming the high quality of the developed QSAR models. The obtained R² and R²ext values for the models are significant and acceptable. The other statistical parameters are also considerable proving the robustness and effectiveness of the reported models. These models can provide mechanistic interpretations and can be the starting point to discover the best drugs for the diseases caused by the irregularities in the function of the KCNQ genes.