Avoiding drug-induced cardiac arrhythmia is recognized as a major hurdle in the successful development of new drugs. The most common problem is acquired long QT syndrome caused by drugs that block human ether-a-go-go-related-gene (hERG) K(+) channels, delay cardiac repolarization and increase the risk of torsades de pointes arrhythmia (TdP). Not all hERG channel blockers induce TdP because they can also modulate other channels that counteract the hERG channel-mediated effect. However, hERG channel blockade is an important indicator of potential pro-arrhythmic liability. The molecular determinants of hERG channel blockade have been defined using a site-directed mutagenesis approach. Combined with pharmacophore models, knowledge of the drug-binding site of hERG channels will facilitate in silico design efforts to discover drugs that are devoid of this rare, but potentially lethal, side-effect.