Sometimes you just want to get straight to the point. Along those lines, FML-seq (fragmentation at methylated loci and sequencing) cuts to the chase by providing a quick look at differential methylation data without the labor and expense of other approaches.
Granted, some methods like whole-genome bisulfite sequencing (WGBS) tell you where every methylation event is, and others such as DARESOME and Joint-snhmC-seq distinguish between 5mC and 5hmC. But in many cases, that’s TMI—you often just need to know if methylation is different between two or more samples. So, that’s where FML-seq comes in. Developed by a cutting-edge team from Robert West’s lab (Stanford), this method combines a methylation-dependent restriction enzyme, MspJI, with special sequencing adapters to determine differential methylation in just three steps. Here are the bare-bones details described by Foley et al.:
- MspJI cuts the DNA a fixed distance from methylated sequences with its motif (mCNNR), generating fragments with a short overhang
- Two things happen in the second step:
- Stem-loop adapters with random overhangs ligate to the fragments, but because they lack a 5’ phosphate, there’s a nick so that DNA polymerase fills in the second strand
- PCR is performed with indexing primers to generate a library
- Fragments are purified and then undergo paired-end sequencing
A few FML-seq advantages that help you escape unnecessary drama are:
- FML-seq doesn’t modify sequences, so they can be easily aligned
- The approach is quick and easy and uses 96-well plates
- Noise is low because the adapters don’t stick to each other
- Unlike similar methods based on MspJI, fragments are purified on beads instead of by inefficient and laborious gel excision
- Costs are further reduced because FML-seq relies on counting sequence reads as opposed to counting bases
When the team compared FML-seq with other similar methods using human cell lines, it did well for both absolute and differential metylation. Methods such as WGBS that generate two channels of data (i.e., methylated and unmethylated), were most similar to each other, as expected. But FML-seq data quality was closer to the data from two-channel approaches than data from single-channel methods like methylated DNA immunoprecipitation (MeDIP-seq). FML-seq is already inexpensive, but with sequencing costs declining, it could be even more attractive as a way to monitor disease progression in the clinic.
“FML-seq is the missing tool for large-scale studies of differential DNA methylation, and perhaps someday for routine clinical testing,” says first author Joe Foley, who conducted the study while in the West lab and is now at the National University of Singapore.
Cut to the chase at Life Science Alliance, September 2023.