Alterations in mitochondrial structure or function have been described in a variety of human diseases for nearly half a century. The complete sequence of the human mitochondrial genome has been published in 1981. The mitochondrial proteome database however, is still incomplete. Here I give a short review on recent advances to determine the complete set of mitochondrial proteins. The main emphasis is put on gel-based proteomic approaches to identify differentially expressed mitochondrial proteins in neurodegenerative diseases.