BenchMasteryMolecular biology techniques & virtual lab

Methylation sequencing

Whole-genome bisulfite sequencing: DNA is bisulfite-converted (before or after adapter ligation), amplified, sequenced deeply and aligned with a conversion-aware aligner to call every CpG.

% methylationSimulator readyCore simulatorEngine: Sequencing & library prep
Choose a sub-technique

Our ruleSimulations prepare you for the bench but don't replace hands-on practice.

Sub-technique you chose

WGBS

Whole-genome bisulfite sequencing: DNA is bisulfite-converted (before or after adapter ligation), amplified, sequenced deeply and aligned with a conversion-aware aligner to call every CpG.

How it differs from the standard method

  • Covers ~28 million CpGs in a human genome at ~30× depth.
  • Bisulfite degrades most of the input, so libraries from small samples are duplicate-rich.
  • GC-rich regions are under-covered; read ends need M-bias clipping.
  • Reads 5mC and 5hmC together.
In this lesson · about 2 min read
  1. Turning a methyl group into a letter
  2. What can go wrong
  3. Choosing a method
  4. Key terms
  5. When to use it
  6. Sub-techniques

1Turning a methyl group into a letter

5-methylcytosine pairs exactly like cytosine, so ordinary sequencing cannot see it. Methylation sequencing makes the difference readable in one of two ways.

WGBS

Whole-genome bisulfite

Every CpG in the genome, but bisulfite fragments most of the DNA and under-covers GC-rich islands.

RRBS

Reduced representation

MspI digestion and size selection keep CpG-dense fragments: a few million CpGs, mostly islands and promoters, cheaply and deeply.

Native

Direct detection

Long native reads report 5mC (and 5hmC) on each molecule, phased to each parental allele.

2What can go wrong

In a conversion method, the reading depends on the chemistry. If conversion is incomplete, unmethylated cytosines stay C and are read as methylated, so every level is inflated, most visibly in unmethylated CpG islands.

3Choosing a method

  • WGBS or EM-seq for unbiased, genome-wide maps including enhancers, gene bodies and partially methylated domains; EM-seq when input is limited or GC-rich regions matter.
  • RRBS for many samples where islands and promoters are the question.
  • Nanopore (or other single-molecule) detection for allele-specific and haplotype-resolved methylation, repeats, and separating 5mC from 5hmC.
  • Conversion methods cannot tell 5mC from 5hmC; oxidative or enzymatic variants that do exist, and native detection reports them separately.
  • Amplification erases methylation: any PCR before conversion, or a PCR-based library for native detection, destroys the signal.

Key terms

Practise these as flashcards →
Bisulfite conversion

Chemical deamination of unmethylated cytosine to uracil in single-stranded DNA; 5mC and 5hmC react much more slowly and remain C.

Conversion efficiency

Fraction of unmethylated C converted, measured on an unmethylated spike-in such as lambda DNA or on non-CpG cytosines.

Beta value

Methylation level at a CpG: methylated reads ÷ all reads covering it, from 0 to 1.

M-bias

Methylation level plotted by position in the read; deviations at read ends reveal end-repair or adapter artefacts to clip.

CpG island

A CpG-rich stretch, often at promoters, usually unmethylated in normal tissue; MspI cuts it into short fragments.

DMR

Differentially methylated region: a stretch whose methylation differs between samples, conditions or alleles.

Imprinting

Methylation of one parental allele only at specific loci, giving ~50% in pooled data but 0% and 100% per allele.

5hmC

5-hydroxymethylcytosine, an oxidised form of 5mC made by TET enzymes; read as C by bisulfite and EM-seq.

When to use it

  • Base-resolution methylation maps for development, cancer and ageing studies.
  • Cohort screens of promoter and island methylation at modest cost (RRBS).
  • Allele-specific methylation at imprinted loci, X inactivation or repeat expansions (long native reads).
  • Biomarker discovery before designing targeted methylation assays.

Compared with alternatives

DNA methylation analysis →

Targeted single-locus methods (methylation-specific PCR, pyrosequencing, MeDIP) for a few regions, cheaply and quickly.

Long-read sequencing →

The platform side of native detection: read length, accuracy and loading decide phasing and coverage.

Whole-genome sequencing →

The same coverage and duplicate logic applies; converted genomes need about the same depth as a 30× germline genome.

Sub-techniques

Choosing one here switches the whole page, including the protocol and the lab, to it.

Whole-genome bisulfite sequencing: DNA is bisulfite-converted (before or after adapter ligation), amplified, sequenced deeply and aligned with a conversion-aware aligner to call every CpG.

  • Covers ~28 million CpGs in a human genome at ~30× depth.
  • Bisulfite degrades most of the input, so libraries from small samples are duplicate-rich.
  • GC-rich regions are under-covered; read ends need M-bias clipping.
  • Reads 5mC and 5hmC together.