This project focuses on understanding photoinduced processes in molecular and biomolecular systems through large-scale quantum molecular dynamics simulations and calculations of time-resolved experimental observables. The overarching goal is to connect nonadiabatic excited-state dynamics with experimentally measurable signatures and thereby obtain a mechanistic understanding of ultrafast photochemical processes. The project combines simulations of electronic and nuclear dynamics with calculations of spectroscopic and diffraction observables, and several parts of the work are carried out in close collaboration with experimental groups. A major ongoing direction concerns photoactive proteins. Building on our previous studies of photoswitchable fluorescent proteins, we are continuing to investigate the photoreactivity of Dronpa2 and rsKiiro, with particular emphasis on the regions where different electronic states become nearly degenerate and photochemical branching occurs. The aim is to understand how molecular structure and the surrounding protein environment control the accessibility of these crossing seams and determine the subsequent reaction pathways. These calculations require extensive sampling of protein configurations together with excited-state electronic-structure calculations and nonadiabatic dynamics. A second major direction investigates chirality-induced spin selectivity (CISS) in photoinduced one- and two-electron transfer. In collaboration with experimentalists, we study chiral donor–bridge–acceptor molecules in which photoexcitation can lead either to charge transfer, involving transfer of a single electron, or to triplet excitation energy transfer through a two-electron mechanism. Using nonadiabatic dynamics and electronic-structure calculations, we aim to determine whether molecular chirality can bias the spin dynamics of these processes and thereby produce spin filtering or spin-polarized photoproducts. Of particular interest is the formation of radical-pair and triplet states, which can be probed experimentally using ultrafast spectroscopy and electron paramagnetic resonance. The simulations are intended both to elucidate the microscopic origin of possible CISS effects and to identify experimentally accessible signatures of spin-selective dynamics. In parallel, we are investigating the ultrafast S2 photoexcited dynamics of acrolein in collaboration with experimental partners. Here, nonadiabatic dynamics simulations are combined with calculations of time-resolved ultrafast electron diffraction signals. By directly simulating the structural observables measured experimentally, we aim to disentangle competing excited-state pathways and identify the nuclear motions and electronic-state transitions responsible for the observed dynamics. Together, these studies require substantial computational resources for ensembles of nonadiabatic trajectories, repeated high-level electronic-structure calculations, sampling of complex molecular environments, and simulation of time-resolved experimental observables. The calculations will provide a direct link between microscopic excited-state dynamics and experimental measurements across photochemistry, photoactive proteins, and molecular spin-dependent processes.