Fig. 2
From: GNINA 1.3: the next increment in molecular docking with deep learning
Covalent docking with Gnina. The input ligand must be provided as conformation representative of the bound form the ligand, including any chemical modifications (e.g. epoxide ring opening). The covalent atom on the ligand is specified with a SMARTS expression; all matching atoms are evaluated. The covalent atom on the receptor is specified with the chain identifier, residue number, and atom name. Additional optional arguments refine the positions and treatment of the covalent bond