Detection and prediction of the onset of seizures are among the most challenging problems in epilepsy diagnostics and treatment. Small electronic devices capable of doing that will improve the quality of life for epilepsy patients while also open new opportunities for pharmacological intervention. This paper presents a novel approach using machine learning techniques to detect seizures onset using intracranial electroencephalography (EEG) signals. The proposed approach was tested on intracranial EEG data recorded in rats with pilocarpine model of temporal lobe epilepsy. A principal component analysis was applied for feature selection before using a support vector machine for the detection of seizures. Hjorth’s parameters and Daubechies discrete wavelet transform coefficients were found to be the most informative features of EEG data. We found that the support vector machine approach had a classification sensitivity of 90% and a specificity of 74% for detecting ictal episodes. Changing the epoch parameter from one to twenty-one seconds results in changing the redistribution of principal components’ values to 10% but does not affect the classification result. Support vector machines are accessible and convenient methods for classification that have achieved promising classification quality, and are rather lightweight compared to other machine learning methods. So we suggest their future use in mobile devices for early epileptic seizure and preictal episodes detection.
Acid-sensing ion channels (ASICs) play a key role in the perception and response to extracellular acidification changes. These proton-gated cation channels are critical for neuronal functions, like learning and memory, fear, mechanosensation and internal adjustments like synaptic plasticity. Moreover, they play a key role in neuronal degeneration, ischemic neuronal injury, seizure termination, pain-sensing, etc. Functional ASICs are homo or heterotrimers formed with (ASIC1–ASIC3) homologous subunits. ASIC1a, a major ASIC isoform in the central nervous system (CNS), possesses an acidic pocket in the extracellular region, which is a key regulator of channel gating. Growing data suggest that ASIC1a channels are a potential therapeutic target for treating a variety of neurological disorders, including stroke, epilepsy and pain. Many studies were aimed at identifying allosteric modulators of ASIC channels. However, the regulation of ASICs remains poorly understood. Using all available crystal structures, which correspond to different functional states of ASIC1, and a molecular dynamics simulation (MD) protocol, we analyzed the process of channel inactivation. Then we applied a molecular docking procedure to predict the protein conformation suitable for the amiloride binding. To confirm the effect of its sole active blocker against the ASIC1 state transition route we studied the complex with another MD simulation run. Further experiments evaluated various compounds in the Enamine library that emerge with a detectable ASIC inhibitory activity. We performed a detailed analysis of the structural basis of ASIC1a inhibition by amiloride, using a combination of in silico approaches to visualize its interaction with the ion pore in the open state. An artificial activation (otherwise, expansion of the central pore) causes a complex modification of the channel structure, namely its transmembrane domain. The output protein conformations were used as a set of docking models, suitable for a high-throughput virtual screening of the Enamine chemical library. The outcome of the virtual screening was confirmed by electrophysiological assays with the best results shown for three hit compounds.