Figure 3.

The computational workflow employed in the screening campaign. The procedure consisted of several steps aimed to find potential, potent SARS-CoV-2 inhibitors, using increasingly accurate in silico methods. First, we conducted pharmacophore screening of over 15M compounds in LigandScout. Then, we proceeded with docking and molecular mechanics–generalized Born and surface area solvation (MM–GBSA) binding energy calculations in Discovery Studio. The number of compounds obtained from the first phase initially increased from around 88 to 120 thousand due to the ligand preparation step, which included creation of possible multiple ionization states. The best 950 potential PLpro inhibitors are gathered in a database. In the last phase, we evaluated these compounds’ affinity toward human UCH-L1 in Maestro software and obtained 387 potential, selective PLpro inhibitors.