Module 9.2 Epigenomic Sequencing

Question 1

Epigenomic sequencing

layers (4)

Answer 1

1. DNA methylation
2. Histone modifications
3. Chromatin accessibility
4. Long-range interactions

Question 2

Chromatin accessibility

Answer 2

• degree to which nuclear micromolecules are able to physically contact chromatin DNA
• chromatin densely arranged within facultative and constitutive heterochromatin
• histones can be replaced by transcription factors at regulatory loci (enhancers, insulators, transcribed gene bodies)
• Transcription factors dynamically compete with histones and other chromatin binding proteins to modulate nucleosome occupancy and promote local access to DNA
• Nucleosome and linker histone occupancy create accessibility spectrum
• chromatin states: closed -> permissive -> open
• Tracing accessibility changes critical to understand epigenetic regulation of gene expression and cellular status

Question 3

long-range interactions

Answer 3

interactions between regulatory elements across genome

Question 4

Cytosine converted forms

Answer 4

1. Cytosine (C)
2. 5-methylcytosine (5-mC)
3. 5-hydroxymethylcytosine (5-hmC)
4. 5-formylcytosine (5-fC)
5. 5-carboxylcytosine (5-caC)

Question 5

Active demethylation process

Answer 5

• TET family proteins convert 5-mC -> 5-hmC ->5-fC -> 5-caC
• 5-mC: most abundant
• 5-hmC: highly abundant in neurons
• 5-fC & 5-caC: lowest abundance, converted to cytosine via BER enzymes

Question 6

5-hmC

features

Answer 6

• highly abundant in neurons
• related to activating genes
• plays important role in embryonic development and cell differentiation
• disregulation been shown in tumorogenesis

Question 7

DNA methylation analysis

bisulfite conversion

features

Answer 7

• unmethylated cytosine converted to uracil via deamination
• methyl group at carbon 5-position (5-mC & 5-hmC) protected against deamination
• denaturation required: sodium bisulfite can only react with cytosine in ssDNA
• PCR: uracils converted to thymines, 5-mC and 5-hmC unchanged
• compared to reference genome
• single nucleotide resolution info about methylation status of DNA segment

Question 8

post-bisulfite adapter tagging

features

Answer 8

• harsh chemical treatment that damages and nicks DNA
• low library conversion rate, significant loss
• utilize single strand library prep to add sequencing adapters to bisulfite-treated ssDNA
• can salvage fragmented DNA caused by bisulfite treatment to mitigate bisulfite induced loss of sequencing templates
• greatly improved library conversion efficiency
• enabled sequencing with very low sample input eg. single cell

Question 9

bisulfite conversion

post-bisulfite adapter tagging

methods (3)

Answer 9

1. Accel-NGS Methyl-Seq
2. TruSeq DNA methylation
3. SPLAT Splintered adapter tagging

Question 10

bisulfite adapter tagging

Accel-NGS Methyl-Seq

process (4)

Answer 10

1. Sheared dsDNA undergo bisulfite conversion
2. adaptase adds adapters at 3’ end
3. single primer extension amplifies 2nd strand with 1st adapter set
4. 2nd adapter ligated to both strands at other end

Question 11

bisulfite adapter tagging

TruSeq DNA Methylation

process

Answer 11

1. whole-genome dsDNA undergoes bisulfite conversion
2. use random primers to attach adapters to copy of ssDNA

Question 12

methylation detection

SPLAT
Splinted adapter tagging

Answer 12

1. sheared dsDNA undergoes bisulfite conversion
2. use adapters with random bases as anchor points for ligation with ssDNA fragment

Question 13

methylation detection

bisulfite sequencing

benefits and drawbacks

Answer 13

Benefits:
- >99% conversion efficiency of unmodified cytosine to uracil
- original gold standard for mapping 5-mC and 5-hmC

Drawbacks:
- limited read length
- 95% of cytosine in genome converted to T
- 60% AT / 40% GC before treatment -> 75% AT, 20% G & <5% C
- lower mapping efficiency and biased genomic coverage
- makes targeted sequencing more difficult
- unable to distinguish 5mc from 5hmc

Question 14

methylation detection

Enzymatic Methyl-Seq
(EM-Seq)

features

Answer 14

• converts unmodified C to U
• mild enzymatic reaction with less damage
• longer DNA fragments than bisulfite conversion - easier sequencing and mapping
• enables low sample input (100 picograms and single-cell seq)
• creates low complexity genome
• > 96% protection rate of methylated cytosine, 0.6% false positive rate
• unable to distinguish 5mc from 5hmc

Question 15

methylation detection

Enzymatic Methyl-Seq
(EM-Seq)

process

Answer 15

1. TET2 protein & β glycosyl transferase (βGT) converts all 5-mC and 5-hmC to 5-gmC to protect against deamination
2. APOBEC3A deaminates unmethylated cytosine to uracil
3. Unmethylated C = T, methylated C = C

Question 16

methylation detection

TET assisted pyridine borane sequencing
(TAPS)

features

Answer 16

• Directly converts methylated C to T
• novel borane reduction chemistry to convert 5-caC to dihydrouracil (DHU)
• Preserves genomic complexity- only 5% of genomic cytosines are methylated
• higher mapping rate and sequencing quality
• Less destructive chemicals
• Enable lower input DNA
• 96% on 5-mC
• false positive rate ~0.3% on unmethylated cytosine
• not robust system that can be easily reproduced
• unable to distinguish 5mc from 5hmc

Question 17

methylation detection

TET assisted pyridine borane sequencing
(TAPS)

process

Answer 17

1. TET1 oxidation of methylated cytosine to 5-caC
2. borane reduction converts 5-caC to DHU
3. DHU reads as T after PCR amplification

Question 18

5-mC / 5-hmC specific detection

methods (6)

Answer 18

5-mC
1. oxBS-Seq
2. TAPSβ

5-hmC
3. ACE-Seq
4. TAB-Seq
5. hmc-CATCH
6. CAPS+

Question 19

5-mC only detection

TAPSβ

process

Answer 19

1. βGT converts 5-hmC to 5-gmC to protect from conversion
2. TET1 oxidation converts 5-mC only to 5-caC
3. borane reduction converts 5-caC to DHU
4. PCR changes DHU to T

Question 20

Methylation detection

Methylated DNA Immunoprecipitation
(MeDIP)

features

Answer 20

• large scale genome wide affinity based method used to enrich for methylated DNA sequences
• isolate methylated DNA fragments through antibody raised against 5-mC
• DNA fragment size (400-600bp) important for immunoprecipitation efficiency, fragment length bias, and final analysis resolution
• antibody needs more than one 5-MC for efficient binding

Question 21

Methylated DNA Immunoprecipitation
(MeDIP)

process

Answer 21

1. genomic DNA fragmentation (400-600bp) and sequencing adapter ligation
2. Denaturation
3. ssDNA DNA incubated with monoclonal anti 5-MC antibodies
4. Magnetic beads conjugated to anti-mouse IGG used to bind anti 5-MC antibodies (unbound DNA removed in supernatant)
5. antibody-bound DNA library molecules eluted and sequenced

Question 22

Methylated DNA Immunoprecipitation
(MeDIP)

benefits and drawbacks

Answer 22

Benefits
- valuable tool for studying whole genome methylation profiles
- no chemical treatment = minimal damage
- Combining with variant detection enables simultaneous profiling of epigenetics and genetics information in genome
- potentially useful for liquid biopsy

Drawbacks
- provides limited, semi quantitative and low resolution methylation information
- resolution determined by fragment size, usually a few hundred base pairs
- accuracy and reliability heavily rely on specificity and affinity of antibody
- Nonspecific binding or cross reactivity with unmethylated DNA = high noise background or false positive results.
- can’t differentiate 5-mC and 5-hmC

Question 23

immunoprecipitation methylation detection

hmC-Seal

features

Answer 23

• affinity based method for specifically profiling 5-HMC
• selectively labels 5-HMC for efficient, unbiased, and genome wide profiling

Question 24

immunoprecipitation methylation detection

hmC-Seal

process

Answer 24

1. adapter ligation to create sequencing adapter-incorporated DNA fragments
2. fragments treated with βGT enzyme to transfer engineered glusode with azide group to 5-HMC in duplex DNA
3. creates modified cytosine with azide group (N3)
4. biotin tag installed onto azide group using click chemistry
5. 5-hmC-containing DNA fragments with biotin tags efficiently captured from random DNA fragments pool by streptavidin beads
6. regular PCR amplification generates sequencing library

Question 25

Nanopore methylation detection

Answer 25

- Methylated bases have different ionic current signature fluctuation than unmethylated base that can be interpreted by base-calling algorithms - methylation status can be picked up directly from sequencing reads in real time without requiring pre-processing of DNA template with conversion or selective pull down - accuracy dependent upon nanopore, processive enzyme, and bioinformatics methodology - Improved sensitivity of channels allows for increasingly subtle modifications to be detected

Question 26

SMRT Technology methylation detection | features

Answer 26

- Data generated from pulse width and interpulse duration - unique kinetic signatures can theoretically be used to identify epigenetic modifications - RCA allows measurement of both strands of initial dsDNA molecule, enabling identification of hemi- and symmetrically methylated positions - sequencing of longer sequences at higher depths requires robust polymerase

Question 27

# methylation detection 3rd generation sequencing | current drawbacks

Answer 27

- high error rate - high cost - high sample input requirement combining chemical or enzymatic reactions of DNA and RNA modifications with 3rd gen sequencing could still provide solution to enhance their detection abilities

Question 28

# histone modification detection Chromatin immunoprecipitation followed by sequencing (ChIP-Seq) | features

Answer 28

- sequence histone modifications in a genome wide manner * Analyze chromatin-associated protein interactions with DNA - can be used to map global binding sites precisely for any protein of interest * Antibodies target specific histone modifications / TFs/ other chromatin binding proteins * sensitive and specific antibody gives high level of enrichment = easier to detect binding events * poor specificity antibody = high background noise

Question 29

Chromatin immunoprecipitation followed by sequencing (ChIP-Seq) | process

Answer 29

1. DNA binding protein crosslinked to DNA in vivo by treating cells with fixation reagents (eg. formaldehyde) 2. chromatin fragmented and incubated with specific antibodies to enrich for DNA-protein complex containing target histone modifications or specific transcription factors 3. cross links reversed and purified DNA sequenced on any NGS platform to detect genome wide histone modifications or protein binding distributions

Question 30

# chromatin accessibility permissive chromatin

Answer 30

- sufficiently dynamic for transcription factors to initiate sequence-specific accessibility remodeling that leads to open chromatin conformation

Question 31

Chromatin Accessibility quantification | methods (4) and feature

Answer 31

1. DNase-seq 2. ATAC-seq 3. MNase-seq 4. NOMe-seq quantify susceptibility of chromatin to either enzymatic modifications or cleavages of its constituent DNA

Question 32

# Chromatin accessibility methods DNase 1 hypersensitive site sequencing (DNase-seq) | features

Answer 32

- quantifies relative abundance of DNA-sensitive chromatin across genome - uses endonuclease DNase 1 to cleave DNA within accessible chromatin - cleavage greatly reduced at protein binding loci - library prep usually multiple day protocol with hundreds of thousands of cells - capture similar regulatory information to ATAC-Seq

Question 33

DNase 1 hypersensitive site sequencing (DNase-seq) | process

Answer 33

- DNase 1 cleaves anywhere there aren't proteins - Accessible library fragments generated by barcoding each cleavage site after digestion - *single cut*: use type II restriction enzyme for 2nd cut to isolate and barcode each DNase cut sites - *double cut*: size-select for NGS-acceptable short fragments arising from paired cleavage events by DNase 1 within DNase hypersensitive sites - ligate adapters to both fragment types - sequence single-cut or short double-cut fragments

Question 34

Assay for Transposable Accessible Chromatin (ATAC-seq)

Answer 34

- uses hyperreactive transposase Tn5 to simultaneously cleave and ligate sequencing adapters into accessible chromatin regions - selectively amplifies double cleavage events in accessible chromatin - widely adopted partly because robustly identifies accessible chromatin - straightforward and rapidly implemented - amenable to materially limited clinical and primary tissue samples - libraries routinely generated in < 2 hours, with 10,000 to 20,000 cells - capture similar regulatory information to DNase-Seq

Question 35

Micrococcal nuclease (MNase)-seq | features

Answer 35

- specifically cuts linker DNA between nucleosomes without crosslinking - acts both as endonuclease to cleave internucleosomal DNA and as exonuclease to degrade cleavage product not protected by protein - uses endonuclease/exonuclease activity to both cleave and eliminate accessible DNA. - sensitivity of MNase digestion can be used to quantify chromatin accessibility - As nucleosome DNA is cleaved less efficiently than inter nucleosome DNA, has been widely used to isolate fragments that span single nucleosomes

Question 36

Micrococcal nuclease (MNase) digestion | Benefits and drawbacks

Answer 36

**Benefits** - removes linker DNA more efficiently than random sonication - more precise nucleosome / histone modification mapping **Drawbacks** - more pronounced sequence bias than sonication. - chromatin solubility creates bias - may be changes in nucleosome positions and histone modifications during experiment without cross linking

Question 37

Nucleosome Occupancy and Methylome Sequencing (NOMe-seq) | features

Answer 37

- uses GpC methyl transferase to methylate accessible DNA - bisulfite conversion of non-methylated cytosine to uracil then sequenced = single molecule accessibility measurement - de novo methyltransferase generates ectopic methylation at GpC dinucleotides orthogonal to endogenous methylation at CpG dinucleotides common in human and mouse genomes - simultaneously probes accessibility and methylation status of DNA at high resolution due to high frequency of GpC throughout genome - requires comparatively large number of sequencing reads to obtain sufficient depths to determine accessibility levels over whole genome - no enrichment bias and single molecule character creates more quantitative view of chromatin accessibility than DNase-Seq, ATAC-Seq or MNase-seq since can directly determine relative accessibility level of each genomic locus

Question 38

Hi-C sequencing | features

Answer 38

- high throughput genomic and epigenomic sequencing technique that comprehensively detects genome wide chromatin interactions in cell nucleus - measures frequency as an average over a cell population at which two DNA fragments physically associate in 3D space, linking chromosome structure directly to genomic sequence - Combining Hi-C data with other data sets (genome wide maps of methylation, chromatin modifications, and gene expression profiles), enables understanding of functional role of chromatin conformation in genome regulation and stability - reveals overall genomic structure of chromosomes - offers insights into biophysical properties of chromatin - specific long range contacts between distant genomic elements eg. (genes and regulatory elements like enhancers)

Question 39

Hi-C sequencing | process

Answer 39

1. cross link chromatin material using formaldehyde 2. chromatin is solubilized and fragmented 3. interacting loci re-ligated together to create genomic library of chimeric DNA molecules. 4. Relative abundance of chimeric molecules / ligation products correlated to probability that respective chromatin fragments interact in 3D space across cell population

Question 40

Cancer epigenetics

Answer 40

- global DNA hypomethylation often accompanied by hypermethylation in specific regions containing CpG islands - hypomethylation induces expression of oncogenes - hypermethylation suppress expression of tumor suppressor genes - occurs early in cancer development - disturbance of other epigenetic mechanisms such as histone modification, DNA binding proteins, non-coding regulatory RNAs observed in different types of cancers - can occur before cancer genetic mutations are detectable - epigenetic changes in cancer cells >> genetic mutations - epigenetic signal contains tissue specific information - promising biomarkers for cancer early screen - potential therapeutic targets