Erratum: Building specificity with nonspecific RNA-binding proteins

Nat. Struct. Mol. Biol. 12, 645–670 (2005).

A mistake was introduced during production in the next-to-last sentence in the legend of Figure 3 (page 649) of the manuscript. The correct legend for Figure 3 is printed here. We apologize for any inconvenience this may have caused. “Figure 3 Functions of the KH and RRM domains. (a) The KH domain binds single-stranded nucleic acid; only the type I fold is shown¹¹³. (b) An RRM domain can interact with a target RNA (i) or protein (ii) via its β-sheet surface. An RRM can also interact with a protein via its α-helical surface on the opposite side (iii). In U2AF65, an extra helix (α3) at the C terminus interacts with the β-sheet surface of RRM3 (iv), which would occlude RNA binding. However, the α3 helix swings away upon RNA binding, facilitating protein dimerization and RNA binding, in U1A RRM, or unfolds upon RNA binding in CstF64 RRM¹¹⁴ (v). Protein names denote examples of each type of interaction.”