Plasmepsins (PMs) are essential proteases of the plasmodia parasites and are therefore promising targets for developing drugs against malaria. We have discovered six inhibitors of PM II by high-throughput fragment-based docking of a diversity set of approximately 40,000 molecules, and consensus scoring with force field energy functions. Using the common scaffold of the three most active inhibitors (IC(50)=2-5 microM), another seven inhibitors were identified by substructure search. Furthermore, these 13 inhibitors belong to at least three different classes of compounds. The in silico approach was very effective since a total of 13 active compounds were discovered by testing only 59 molecules in an enzymatic assay. This hit rate is about one to two orders of magnitude higher than those reported for medium- and high-throughput screening techniques in vitro. Interestingly, one of the inhibitors identified by docking was halofantrine, an antimalarial drug of unknown mechanism. Explicit water molecular dynamics simulations were used to discriminate between two putative binding modes of halofantrine in PM II.