regions, 5
0 -UTRs, gene bodies, 3
0 -UTRs, and intergenic regions [71]. The Infinium
450 K is also applicable to both cell and tissue samples; however this array lacks
coverage of distal regulatory elements [74].
Although the Infinium 450 K had been widely used by the epigenetics research
community, the technology has presented some technical challenges [74]. It was
reported that many of the probes in the Illumina 450 K array either cross-hybridize to
nontargeted genomic regions or target loci that contain known single-nucleotide
polymorphisms (SNPs), both of which can interfere with analysis of DNA methylation levels [75]. Because of this, sites of differential DNA methylation identified
using the Illumina 450 K array must be confirmed using a secondary independent
assay, especially the sites targeted by the cross-reactive probes or that contain known
SNPs. This led Illumina to release a new technology, the Infinium MethylationEpic
(EPIC) BeadChip. This new array was specifically designed to target enhancer
regions and avoid SNPs and cross-reactive probes. It contains over 850,000 probes,
which cover more than 90% of the sites on the Infinium 450 K, plus more than
350,000 CpGs at regions identified as potential enhancers by FANTOM5 and the
ENCODE project [72, 73]. In this array, probe sequences are designed to be
complementary to specific 50 base regions of bisulfite-converted genomic DNA
with a CpG site at the 3
0 end of the probe. Following hybridization to bisulfiteconverted DNA, single-base extension of the probe incorporates a fluorescently
labeled ddNTP at the 3
0 CpG site to distinguish between the C/T conversion that
results from bisulfite treatment [74].
Methylation array methodologies described above represent a cost-effective and
high-throughput technology to quantify the methylation level of half a million
cytosines all over the genome. Furthermore, these arrays have low input DNA
requirements, making them a powerful tool to study DNA methylomes. However,
there are potential biases with this technology. As with all bisulfite-based methods,
incomplete bisulfite conversion can affect accurate quantitation of methylation
differences. In addition, single-nucleotide polymorphisms (SNPs) can interfere
with methylation level detection, and commercial arrays are limited to human
samples only. No commercial arrays are now available to map methylation of the
murine genome, hindering the application of this methodology to the wide range of
valuable animal models of disease [76, 77]. Despite these limitations, Needhamsen
et al. recently reported the use of the Infinium Human MethylationEPIC BeadChip to
study DNA methylation in mouse DNA samples [77]. This study identified over
19,000 EPIC probes which aligned to the bisulfite-converted reference mouse
genome [77]. Another limitation of this technology is its inability to differentiate
methylation from hydroxymethylation [78].
Traditional array methods do not distinguish between 5mC and 5hmC as both are
resistant to deamination. However, recent work by Nazor et al. combined
TET-assisted bisulfite (TAB) conversion with Illumina 450 K DNA methylation
arrays. This method, termed TAB-array, provides a low-cost, high-throughput
approach that distinguishes 5hmC from 5mC signals at base resolution [79]. This
technique requires that the genomic DNA from each biological sample is split in two
fractions for different downstream processing steps. One fraction undergoes normal
502
C. L. Seiler et al.
0 -UTRs, gene bodies, 3
0 -UTRs, and intergenic regions [71]. The Infinium
450 K is also applicable to both cell and tissue samples; however this array lacks
coverage of distal regulatory elements [74].
Although the Infinium 450 K had been widely used by the epigenetics research
community, the technology has presented some technical challenges [74]. It was
reported that many of the probes in the Illumina 450 K array either cross-hybridize to
nontargeted genomic regions or target loci that contain known single-nucleotide
polymorphisms (SNPs), both of which can interfere with analysis of DNA methylation levels [75]. Because of this, sites of differential DNA methylation identified
using the Illumina 450 K array must be confirmed using a secondary independent
assay, especially the sites targeted by the cross-reactive probes or that contain known
SNPs. This led Illumina to release a new technology, the Infinium MethylationEpic
(EPIC) BeadChip. This new array was specifically designed to target enhancer
regions and avoid SNPs and cross-reactive probes. It contains over 850,000 probes,
which cover more than 90% of the sites on the Infinium 450 K, plus more than
350,000 CpGs at regions identified as potential enhancers by FANTOM5 and the
ENCODE project [72, 73]. In this array, probe sequences are designed to be
complementary to specific 50 base regions of bisulfite-converted genomic DNA
with a CpG site at the 3
0 end of the probe. Following hybridization to bisulfiteconverted DNA, single-base extension of the probe incorporates a fluorescently
labeled ddNTP at the 3
0 CpG site to distinguish between the C/T conversion that
results from bisulfite treatment [74].
Methylation array methodologies described above represent a cost-effective and
high-throughput technology to quantify the methylation level of half a million
cytosines all over the genome. Furthermore, these arrays have low input DNA
requirements, making them a powerful tool to study DNA methylomes. However,
there are potential biases with this technology. As with all bisulfite-based methods,
incomplete bisulfite conversion can affect accurate quantitation of methylation
differences. In addition, single-nucleotide polymorphisms (SNPs) can interfere
with methylation level detection, and commercial arrays are limited to human
samples only. No commercial arrays are now available to map methylation of the
murine genome, hindering the application of this methodology to the wide range of
valuable animal models of disease [76, 77]. Despite these limitations, Needhamsen
et al. recently reported the use of the Infinium Human MethylationEPIC BeadChip to
study DNA methylation in mouse DNA samples [77]. This study identified over
19,000 EPIC probes which aligned to the bisulfite-converted reference mouse
genome [77]. Another limitation of this technology is its inability to differentiate
methylation from hydroxymethylation [78].
Traditional array methods do not distinguish between 5mC and 5hmC as both are
resistant to deamination. However, recent work by Nazor et al. combined
TET-assisted bisulfite (TAB) conversion with Illumina 450 K DNA methylation
arrays. This method, termed TAB-array, provides a low-cost, high-throughput
approach that distinguishes 5hmC from 5mC signals at base resolution [79]. This
technique requires that the genomic DNA from each biological sample is split in two
fractions for different downstream processing steps. One fraction undergoes normal
502
C. L. Seiler et al.