Global minimization methods and their applications to structural molecular biology have made significant progresses in recent years. Appropriately simplified molecular docking problem is a good object of global minimization, which is now a rather active research area. Molecular docking can be divided into two categories. The detailed docking for de novo ligand design and the rough docking for known chemical database screening for drug discovery. Their demands for the global minimization algorithms are different. New stochastic and deterministic global minimization algorithms that are suitable for docking problems are briefly reviewed. Those algorithms with potential smoothing techniques seem to be promising and are worth close noting.