Researchers learn through experience not to take scientific facts for granted. Indeed, every so often, well-established views suffer a battering. Take protein folding, for example. In the traditional view, proteins need to be folded into a highly specific three-dimensional structure for them to function properly. But, as H. Jane Dyson and Peter E. Wright discuss in a Review article on page 197, the occurrence of unstructured regions in functional proteins is, in fact, surprisingly common, and disordered regions can be highly conserved in homologous proteins of different species. In addition, recent technical advances have led to the realization that intrinsically disordered proteins are involved in important cellular processes. So, what sets this class of proteins apart and how do they function? The characterization of these proteins has only just begun, but it is already becoming apparent that our concept of a functional protein must evolve from a static picture of a folded protein to a highly dynamic picture, in which numerous structural conformations that are consistent with various aspects of protein function are represented.

In another example — the cell-death field — the dogma used to be that a class of cysteine proteases, known as caspases, trigger processes that ultimately cause apoptotic cell death. However, accumulating data seem to indicate that caspases aren't necessarily required for cells that have been treated with an apoptotic inducer to become committed to dying. The evidence for caspase-independent cell death is outlined in an Opinion article on page 268 by Jerry E. Chipuk and Douglas R. Green. Having uncovered a possible new type of cell death, the next challenge will be to understand the underlying mechanisms and its role in physiological cell death.