Figuring out human DNA methylation patterns with bisulfite sequencing can be tough, but the complexities of plant DNA methylation can turn your lab into a Little Shop of Horrors. That’s because plants methylate cytosine residues not only in the CG sequence context, but also in CHG and CHH (where H is any nucleotide but guanine). In a recent article in Epigenetics, researchers from the US, UK, and China describe potential pitfalls of sodium bisulfite sequencing in plants and share helpful tips:
- Problem: Amplification of un-bisulfite-converted genomic DNA makes methylated C’s appear to sprout like weeds.
- Solution: Use more stringent bisulfite conversion protocols to fully convert unmethylated C’s to U’s. Tweak PCR primer design to preferentially amplify fully bisulfite-converted DNA.
- Problem: Sibling clones, which occur when the same DNA molecule is sequenced over and over again, cause overrepresentation of particular methylation patterns.
- Solution: Recognize sibling clones as sequencing reads with identical CHH methylation patterns (CHH sites are methylated at a low frequency, so the odds of two independent clones having the same methylated CHH distribution are unlikely). Include only one clone with a given CHH methylation pattern in the analysis.
Although especially important for bisulfite sequencing of plant DNA, similar principles apply to the analysis of any DNA sample. For more details, check out Epigenetics, January 2010.