“…To build our model, we have relied on distinct in silico approaches. Methodological choices are closely based on the rationale adopted in a similar study by one of the authors on enzymatic stereoselectivity and are in line with options suggested in several reviews referenced earlier. − , In the Supporting Information, we evaluate the reliability of our choices, focusing on the following aspects: (1) the suitability of both chosen MD approaches to evaluate % reac and Δ F ⧧ and their statistical validity, (2) the suitability of the semiempirical QM method DFTB3/O-Phyd in evaluating Δ F ⧧ , especially in relation to other methodological options ,,− ,,,,,,,, and whether benchmarking is satisfactory, and (3) how well sampling along the relatively simple RC Enz and RC Sub still manages to capture chemical changes involving other QM atoms.…”