The present thesis addresses different areas of Computer-Aided Drug Design: 1) Binding affinity prediction: Steered molecular dynamics allowed to distinguish affinity differences as small as 4.3 μM for substrate analogues of the Herpes Simplex Virus 1 Thymidine Kinase. 2) Aggregation breakers: Homology modelling of the primary aggregation of the monoclonal antibody bevacizumab as well as docking and scaffold hopping led to the identification of AMP as a promising antibody aggregation breaker. 3) Protein-protein interaction inhibitors: Virtual screening yielded 43 potential inhibitors of a protein-protein interface of the NADPH oxidase 2 (NOX2), docking led to a binding mode for celastrol, and a GRID-based interaction analysis identified peptides with binding affinities comparable to the native peptide. 4) Target discovery: Potential targets for phenazopyridine were identified by inverse in silico screening, of which 80% were confirmed experimentally.