Machine Learning Optimization of non-Kasha Behavior and of Transient Dynamics in Model Retinal Isomerization

TL;DR

This paper uses multi-objective Bayesian optimization to refine a minimal retinal isomerization model, accurately predicting non-Kasha fluorescence behavior and transient dynamics in Rhodopsin, aligning closely with experimental data.

Contribution

It introduces a novel application of Bayesian optimization to a minimal model, improving the prediction of transient and steady-state behaviors in retinal isomerization.

Findings

Accurate prediction of excitation wavelength-dependent fluorescence spectra.

Reduction of discrepancies in transient dynamics.

Excellent agreement with experimental data.

Abstract

Designing a model of retinal isomerization in Rhodopsin, the first step in vision, that accounts for both experimental transient and stationary state observables is challenging. Here, multi-objective Bayesian optimization is employed to refine the parameters of a minimal two-state-two-mode (TM) model describing the photoisomerization of retinal in Rhodopsin. With an appropriate selection of objectives, the optimized retinal model predicts excitation wavelength-dependent fluorescence spectra that closely align with experimentally observed non-Kasha behavior in the non-equilibrium steady state. Further, adjustments to the potential energy surface within the TM model reduce the discrepancies across the time domain. Overall, agreement with experimental data is excellent.