Opsins are light-gated ion channels that, upon light absorption at specific wavelengths, respond by opening or closing, conducting the flow of ions into or out of the neurons, resulting in their activation or inhibition. The system of study in this project will be Channelrhodopsin-2 (ChR2), a subfamily of microbial-type rhodopsins [1]. It consists of seven transmembrane helices, with a covalently bound retinal as chromophore [2]. Light absorption is followed by the photoisomerization of the all-trans retinal to the 13-cis configuration, driving cyclic conformational changes of the molecule, namely a photocycle, which consists of several intermediates. This project is a continuation of a previous NAISS Small Compute and Storage allocation that successfully enabled us to publish results in Ref. [4]. In that study, we performed simulations using recently developed quantum mechanics (QM)/molecular mechanics (MM) advanced methods on the ChR2 embedded in a lipid membrane. Thanks to that Compute allocation, we documented in detail how to treat each component of a complex biomolecular environment—including the protein dimer, lipid bilayer, solvent, and ions—using fragment-based approaches and averaged parameters in a consistent, practical workflow. That workflow consisted of classical MD, QM/MM MD, fragment-based polarizable embedding to build environment-specific PE potentials and PE-TD-DFT for the spectroscopic calculations. We obtained an excellent agreement with experimental one-photon absorption spectra. Additionally, to our knowledge, we also provided the first theoretical two- and three-photon absorption spectra for ChR2, in the absence of direct experimental data. As a continuation, we would like to elucidate the complete two-photocycle model of Channelrhodopsin 2, as proposed experimentally in Ref. [3], including the seven aforementioned intermediates, using the published workflow [4]. Most of the steps of this photocycle are not completely understood, and theoretical calculations are needed to drive new experiments. Additionally, we would like to use enhanced sampling simulations to reach metastable conformations of the system that would otherwise be inaccessible via classical MD. Refs: [1] Science 296, 2395–2398 (2002) [2] Nature 482, 369–374 (2012) [3] Bio. and Physicobio., 2017, Vol. 14, 13−22. [4] J. Chem. Theory Comput. 2026, https://doi.org/10.1021/acs.jctc.5c01719