DNA double helix with methylcytosines.

The often-quoted 'four bases' that make up the code of life are really five. The occurrence of a methyl modification on cytosine has important implications for the regulation of DNA transcription; abnormal methylation can wreak havoc with gene regulation and can be the underlying cause of disease.

Given its importance, methods to systematically map the methylome—that is, all genome-wide occurrences of methylcytosine—have recently flourished. Joseph Ecker and colleagues just provided a complete map of all methylcytosines in human embryonic stem cells at base-pair resolution (Nature 462, 315–322, 2009). What this and other sequencing-based methods, such as the strategy developed by Meissner et al. that uses reduced representation libraries (Nature 454, 766–770, 2008), have in common is that they rely on bisulfite conversion, the chemical conversion of unmethylated cytosine to uracil that is then converted to thymine in subsequent amplification steps. Though effective, this approach requires some hands-on time, to ensure complete conversion, and computational effort to map sequences using a redundant code in which a thymine can represent either a thymine or an unmethylated cytosine in the original sequence.

Earlier this year, a report by scientists at Oxford Nanopore Technologies (Nat. Nanotechnol. 4, 265–270, 2009) held out the tantalizing possibility that this labor-intensive technique could be replaced by a simple sequencing step with a sequencer that can directly distinguish unmodified cytosine from methylcytosine. In a nanopore, each base that traverses through it is identified by the current amplitude, the extent to which it blocks the current that is flowing through the pore. Methylcytosine produces a signal clearly distinguishable from those of the other four bases and can thus be read directly.

At present nanopores have only been shown to sequence short oligonucleotides, a far cry from the whole-genome sequencing achieved by bisulfite conversion, and some technical hurdles, such as making sure each base enters the nanopore in the right order and gets swept out on the other side, still need to be worked out—but once they are, the direct sequencing of the fifth base will have a big impact on the field. Let's see if it happens in 2010.