Excited states of the green fluorescent protein chromophore: Performance of ab initio and semi-empirical methods

The complex photophysical properties of fluorescent proteins give rise to a wide field of applications as markers in molecular biology. Understanding these properties is essential for rational genetic engineering of new fluorescent proteins. Here, theoretical models are required to support the interpretation of structural and spectroscopic experimental data. This requires the accurate and reliable prediction of excited‐state features of the chromophore and its interactions with the protein matrix. Here, we compare calculations of absorption and emission energies of semi‐empirical (OM2/MRCI, ZINDO/S, and TD‐DFTB) and ab initio (SORCI, CC2, and TDDFT) approaches for the HBDI chromophore in vacuo and wild‐type green fluorescent protein (GFP) using QM/MM models. We discuss the influence of electrostatic fields, the chromophore geometry, the size of the QM region, and methodological aspects, in particular charge‐transfer states in TDDFT and the applicability of real‐space TDDFT codes. We revisit previous opposing theoretical studies and benchmark gas‐phase measurements. Proton transfer wire of wild‐type green fluorescent protein (wt‐GFP).