Methylated cytosines can be detected at single-base resolution by treating DNA with sodium bisulfite and then sequencing. Sodium bisulfite, however, does not discriminate between canonical 5-methylcytosine (5mC) and the recently discovered variant 5-hydroxymethylcytosine (5hmC), which may have unique epigenetic functions. Two groups have described methods that selectively change the chemical properties of one form of modified cytosine but not the other. Yu et al. protected 5hmC by enzymatic modification and then oxidized 5mC with an enzyme called TET1. In contrast, Booth et al. selectively oxidized 5hmC using a chemical, potassium perruthenate. In both methods, the samples were then treated with sodium bisulfite, sequenced and the resulting data were compared with data from a standard bisulfite sequencing experiment to quantitatively assess the distribution and abundance of both 5hmC and 5mC. Yu et al. performed a genome-wide analysis in both human and mouse embryonic stem cells, finding ∼690,000 5hmC residues in human cells and ∼2 million in mouse at a 5% false-discovery rate. Booth et al. analyzed CpG islands in mouse embryonic stem cells, finding 800 CpG islands containing 5hmC compared with ∼4,500 CpG islands with 5mC at a 3.7% false-discovery rate. The studies reveal that 5hmC is enriched at enhancer elements and that 5hmC shows an asymmetric strand bias toward G-rich sequences. (Science 336, 934–937, 2012; Cell published online, doi:10.1016/j.cell.2012.04.027, 17 May 2012)