The methylation state of DNA (specifically that of the base cytosine) has been shown to influence the expression of genes. For example, in mammalian cells higher levels of methyl CpG around the transcription start site (TSS) have been associated with transcriptional silencing, although more complex patterns in other regions of the genome are being revealed. The rise of next generation sequencing has enabled the transition from studying the methylation patterns of just a few genes or small regions to that of truly genome-wide studies, and this is leading to a far richer view of the methylome.
There are a variety of methods for monitoring the methylation status of the genome, but they can generally be placed into one of two categories – they either rely on bisulfite conversion or they employ a form of methylated DNA enrichment or pulldown. Bisulfite treatment of DNA converts all unmethylated cytosines to uracil (which is read as thymine) while 5′-methyl-cytosine is left unchanged. Currently the most complete picture of the methylome is generated via whole genome bisulfite sequencing (WGBS). Unfortunately, it is also the most expensive at around 1.5X the cost of standard whole genome sequencing. In an effort to reduce the cost and to increase the sample throughput, several methods have been developed which limit sequencing to only a portion of the genome. One such example is “reduced representation bisulfite sequencing” (RRBS) which uses restriction enzymes and size selection to reduce the overall complexity of the genome while enriching for CpG islands (regions of high CpG density). It should be noted, however, that this enrichment introduces some biases as to which CpG sites end up being sequenced.
Another strategy for lowering the costs and increasing the throughput of methylation sequencing is to select for methylated DNA prior to the sequencing step. One common method is “methylated-DNA immunoprecipitation” (MeDIP-Seq). Similar to ChIP-Seq, MeDIP-Seq is performed by immunoprecipitating methylated DNA with an antibody raised against 5′-methylcytosine. The unmethylated DNA is washed away, leaving the material highly enriched for methylated DNA. The presence or absence of a particular sequence gives an estimate of the level of methylation in that region of the genome. A similar method, called “methyl CpG immunoprecipitation” (MCIp), uses a methyl-CpG binding domain (MBD) protein to isolate the methylated regions of the genome. While offering genome-wide coverage, both of these methods introduce some level of bias and data interpretation can be tricky.
In general, methylation sequencing applications are most suitable for those platforms which generate a large amount of sequence per run. For example, WGBS requires about 1.5X the amount of sequence needed for standard genome sequencing. The reduced complexity methods require less sequencing, but the demands are still fairly high, with human samples requiring approximately 5-10 Gb per sample. For bisulfite converted samples, it should be noted that the reduced genomic complexity (with all unmethylated C’s being converted to T’s) can create alignment challenges. Finally, the ideal platform would be one that is able to read the methylation status of the DNA directly (without the need for bisulfite conversion). While preliminary proof of concept studies along these lines have been performed on the PacBio RS, it has not yet been transformed into a commercially viable application.
Please contact us at firstname.lastname@example.org if you have any information or opinions you’d like to share about this page.