Proteins with large and flat binding sites as well as protein–protein interactions are considered ' undruggable ' with conventional small-molecule drugs. Cyclic peptides have been found to be capable of binding to such targets with high affinity, making this class of compounds
an interesting source for possible therapeutics. However, the oftentimes poor passive membrane permeability of cyclic peptides still imposes restrictions on the applicability of cyclic peptide drugs. Here, we describe how computational methods in combination with experimental data can be used
to improve our understanding of the structure–permeability relationship. Especially the conformational dynamic and chameleonic nature of cyclic peptides, which we investigate by a combination of MD simulations and kinetic modeling, is important for their ability to permeate passively
through the membrane. The insights from such studies may enable the formulation of design principles for the rational design of permeable cyclic peptides.