Discovering a new use for something is always a nice suprise, sort of like finding a forgotten $20 bill in your back pocket. Researchers at Pacific Biosciences realized that their single-molecule, real-time (SMRT) sequencing technique could also have the bonus ability to directly detect DNA methylation (DNAm)—without bisulfite conversion or any other additional sample prep.
With SMRT sequencing, DNA polymerase strings together fluorescently labeled nucleotides complementary to the template sequence. The team already knew that measuring the duration and spacing of the fluorescent pulses reveals info about the primary and secondary DNA structure. They figured that these data also could tell them about DNAm. Turns out they were right.
A few more details:
- Compared with control unmethylated templates, the interpulse durations (IPDs) of methylated synthetic templates were very different at and near methylated bases.
- mC and hmC mods had unique kinetic values in particular sequence contexts on the synthetic templates. Combining this info with statistical methods, the team could distinguish between cytosine, mC, and hmC, something that you just can’t do with bisulfite sequencing.
- Multiple reads (“circular consensus sequencing”) and statistics help in cases where you have variably methylated genomic sites.
- The long read lengths will probably help map DNAm in repetitive genomic regions, where there’s usually a lot of mC.
Although genomic mA detection was a piece of cake, genomic mC and hmC detection with base-pair resolution will still require a bit of tweaking to get it just right, say the researchers.
Get even SMRT-er by reading the full paper at Nature Methods, May 2010.