A Deep Learning\nApproach to Uncover Voltage-Gated\nIon Channels’ Intermediate States

Owing\nto recent advancements in cryo-electron microscopy,\nvoltage-gated\nion channels have gained a greater comprehension of their structural\ncharacteristics. However, a significant enigma remains unsolved for\na large majority of these channels: their gating mechanism. This mechanism,\nwhich encompasses the conformational changes between open and closed\nstates, is pivotal to their proper functioning. Beyond the binary\nstates of open and closed, an ensemble of intermediate states defines\nthe transition path in-between. Due to the lack of experimental data,\none might resort to molecular dynamics simulations as an alternative\nto decipher these states and the transitions between them. However,\nthe high-energy barriers and the colossal time scales involved hinder\naccess to the latter. We present here an application of deep learning\nas a reliable pipeline for a comprehensive exploration of voltage-gated\nion channel conformational rearrangements during gating. We showcase\nthe pipeline performance specifically on the Kv1.2 voltage sensor\ndomain and confront the results with existing data. We demonstrate\nhow our physics-based deep learning approach contributes to the theoretical\nunderstanding of these channels and how it might provide further insights\ninto the exploration of channelopathies.