*2.2. Geometry Optimization–Complexes*

The search for the optimal geometries for complexes of twenty amino acids (for formulas and abbreviations see the Supplementary Information) with *pc*, *wc*, and *al* conformers of either calix[6]arene, or hexa-*p*-*tert*-butylcalix[6]arene has been performed in several steps. Firstly, the generation of approximate geometries of the most plausible host–guest conformers was performed with the AutoDock Vina program [86]. Secondly, the resulting geometries were used as starting points for the DFT optimization. Because of the high computational cost, the def2-SVP basis set was used for the search of local minima, and solely the lowest ones were reoptimized using the def2-TZVP basis set. It should be noted that for the arginine (Arg) case, all AutoDock Vina optimizations lead to the protonated (charged) version of the amino acid. Since the interaction energies between two ions would give a completely different interaction picture than for uncharged molecules, we decided to skip this amino acid from our test set. For further analysis, only the lowest minima for each calixarene conformer and amino acid were utilized.

Gaussian16 [87] was used for the geometry optimizations. The harmonic frequency analysis has been performed in order to verify that the stationary point is the minimum.
