Application and Comparison of Methylation Snapshot and MPS Methods - - PDF document

▶
application and comparison of methylation snapshot and
SMART_READER_LITE
LIVE PREVIEW

Application and Comparison of Methylation Snapshot and MPS Methods - - PDF document

Aug 31, 2023 493 likes •609 views

10/10/2018 Application and Comparison of Methylation Snapshot and MPS Methods to Analyze Epigenetic Age Signatures in Saliva Sae Rom Hong 1,2 , Sang-Eun Jung 1 , Eun Hee Lee 1 , Kyoung-Jin Shin 1,2 , Woo Ick Yang 1 , Hwan Young Lee 1, * 1

slide-1
SLIDE 1

10/10/2018 1

1Department of Forensic Medicine, Yonsei University College of Medicine, Seoul, Korea 2Department of Forensic Medicine and Brain Korea 21 PLUS Project for Medical Science, Yonsei University, Seoul, Korea

Sae Rom Hong1,2, Sang-Eun Jung1, Eun Hee Lee1, Kyoung-Jin Shin1,2, Woo Ick Yang1, Hwan Young Lee1,*

Application and Comparison of Methylation Snapshot and MPS Methods to Analyze Epigenetic Age Signatures in Saliva

DNA Methylation

  • Addition of a methyl group to cytosine followed by guanine
  • 5’-CG-3’
slide-2
SLIDE 2

10/10/2018 2

DNA Methylation

Cell differentiation Genetic factor Environmental factor

[ ]

Body Fluid Identification

Aging

[ ]

Age Prediction

[ ]

Genetic Traits G

[ ]

Behavior Habits

Aging

[ ]

Age Prediction

Method

  • HumanMethylation450 BeadChip Array

‒ 54 males (18-73 years) ‒ Marker candidates selection by multivariate linear regression analysis

  • Targeted Bisulfite Sequencing

‒ Multiplex methylation SNaPshot (226 samples; Both sets) ‒ Massively parallel sequencing (95 samples; Training set) ‒ Analysis using several tools (SPSS, etc.)

  • Saliva samples

‒ 280 samples (18-73 years) Info Training Set Testing Set Total Male 47 70 117 Female 48 61 109 Total 95 131 226

slide-3
SLIDE 3

10/10/2018 3

Marker Selection

Hong et al. FSI Genet. (2017)

  • 6 age-associated CpG candidates

+ Cell type-specific marker (cg18384097) HumanMethylation450 BeadChip Array

Cell Type-specific Marker

R² = 0.9286

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Methylation at cg18384097 Buccal-Cell-Signature (ϐ) (predicted epithelial cell compositions) N = 54 Spearman’s rho = 0.955

  • cg18384097 (PTPN7)

‒ Souren et al. Genome Biol. (2013) ‒ High in buccal epithelial cell ‒ Low in blood cell ‒ PTPN7 gene

  • Protein tyrosine phosphatase (PTP)
  • Preferentially expressed in hematopoietic cells
  • Buccal-Cell-Signature (ϐ)

‒ Eipel et al. Aging (Albany NY). (2016) ‒ cg07380416 (CD6) ‒ cg20837735 (SERPINB5) ‒ Percentage of buccal epithelial cells

ϐ = 99.8 × + 1.92 2 + −98.12× + 88.54 2

slide-4
SLIDE 4

10/10/2018 4

Detail Workflow

Massively Parallel Sequencing (N=95)

Bismark

Multiplex PCR

Read Sequence

Indexing PCR

Index Sequence

10ng

Bisulfite conversed DNA Multiplex Methylation SNaPshot (N=226=95+131)

Multiplex PCR

DNA Methylation Analysis

Multiplex SBE

A G

Hong et al. FSI Genet. (2017) Lee et al. FSI Genet. (2016)

Analysis tools

Methylation SNaPshot (N=95)

R² = 0.5412 0.1 0.2 0.3 0.4 20 40 60 80 Methylation Chronological Age (years)

cg00481951 (SST)

R² = 0.2541 0.1 0.2 0.3 0.4 20 40 60 80 Methylation Chronological Age (years)

cg19671120 (CNGA3)

R² = 0.6313 0.1 0.2 0.3 0.4 20 40 60 80 Methylation Chronological Age (years)

cg14361627 (KLF14)

R² = 0.4193 0.2 0.4 0.6 0.8 1 20 40 60 80 Methylation Chronological Age (years)

cg08928145 (TSSK6)

R² = 0.2139 0.1 0.2 0.3 0.4 0.5 0.6 20 40 60 80 Methylation Chronological Age (years)

cg12757011 (TBR1)

R² = 0.5928 0.1 0.2 0.3 0.4 0.5 0.6 20 40 60 80 Methylation Chronological Age (years)

cg07547549 (SLC12A5)

slide-5
SLIDE 5

10/10/2018 5

Methylation SNaPshot (N=226)

R² = 0.4791 0.1 0.2 0.3 0.4 20 40 60 80 Methylation Chronological Age (years)

cg00481951 (SST)

R² = 0.2882 0.1 0.2 0.3 0.4 20 40 60 80 Methylation Chronological Age (years)

cg19671120 (CNGA3)

R² = 0.6347 0.1 0.2 0.3 0.4 20 40 60 80 Methylation Chronological Age (years)

cg14361627 (KLF14)

R² = 0.4341 0.2 0.4 0.6 0.8 1 20 40 60 80 Methylation Chronological Age (years)

cg08928145 (TSSK6)

R² = 0.167 0.1 0.2 0.3 0.4 0.5 0.6 20 40 60 80 Methylation Chronological Age (years)

cg12757011 (TBR1)

R² = 0.5486 0.1 0.2 0.3 0.4 0.5 0.6 20 40 60 80 Methylation Chronological Age (years)

cg07547549 (SLC12A5)

Model – Multiplex Methylation SNaPshot

Target ID Coefficient (intercept)

  • 24.521

cg18384097

  • 31.111

cg00481951

6.718

cg19671120

23.760

cg14361627

81.053

cg08928145

24.325

cg12757011

53.634

cg07547549

89.415

20 40 60 80 20 40 60 80

Predicted Age (years) Chronological Age (years)

Training Set (N=95)

20 40 60 80 20 40 60 80

Predicted Age (years) Chronological Age (years)

Testing Set (N=131) MAD = 3.03 RMSE = 4.03 MAD = 3.43 RMSE = 4.36

MAD: Mean Absolute Deviation RMSE: Root Mean Square Error

slide-6
SLIDE 6

10/10/2018 6

MPS Analysis

  • Platform

‒ MiSeq Reagent Kit v3 (2×300) ‒ HiSeq 2000

  • Pipeline

Trimming Bismark Raw file CpG report Integrated data ‒ Adapter ‒ Quality ‒ Alignment ‒ CpG Extraction

MPS (N=95) – CpG sites in amplicons

ID

CpG 1 CpG 2 CpG 3 CpG 4 CpG 5 CpG 6 CpG 7 CpG 8 CpG 9 CpG 10 CpG 11 CpG 12 CpG 13 CpG 14 CpG 15

cg18384097 -.179

  • .163
  • .162
  • .150
  • .163
  • .180

cg00481951 .682* .799* .814* .501* .421* .311* .381* .478* cg19671120 .187 .067 .104

  • .033

.135 .194 .433* .507* .336* .325* .501* .483* .521* .560* .567* cg14361627 .261* .492* .556* .631* .650* .756* cg08928145 .596* .584* .662* .649* .637* .637* .616* .637* .641* .629* .636* cg12757011 -.035 .229* .319* .489* -.002 cg07547549 .321* .130 .441* .571* .585* .683* .741* .679* .769* .399* .285* Tageted CpG site in the methylation SNaPshot

  • Pearson’s R (Correlation between chonological age and methylation)

*Statistically significant

slide-7
SLIDE 7

10/10/2018 7

MPS (N=95) – Methylation value

R² = 0.3219

0.05 0.1 0.15 0.2 0.25 20 40 60 80

Methylation Chronological Age (years)

cg19671120_CpG15

R² = 0.5714

0.1 0.2 20 40 60 80

Methylation Chronological Age (years)

cg14361627_CpG6

R² = 0.2393 0.25 0.5 20 40 60 80 Methylation Chronological Age (years)

cg12757011_CpG4

R² = 0.5917 0.2 0.4 20 40 60 80 Methylation Chronological Age (years)

cg07547549_CpG9

R² = 0.6626 0.1 0.2 20 40 60 80 Methylation Chronological Age (years)

cg00481961_CpG3

R² = 0.4043 0.5 1 20 40 60 80 Methylation Chronological Age (years)

cg08928145_CpG13 Target ID Coefficient (intercept)

  • 24.521

cg18384097

  • 31.111

cg00481951 6.718 cg19671120 23.760 cg14361627 81.053 cg08928145 24.325 cg12757011 53.634 cg07547549 89.415

Age Prediction Using the MPS Data

  • 20

20 40 60 80 20 40 60 80

Predicted Age (years) Chronological Age (years)

SNaPshot model (N=95)

SNaPshot model

MAD = 22.43 RMSE = 24.13

slide-8
SLIDE 8

10/10/2018 8

Methylation SNaPshot vs MPS (N=95)

G (Methylated) A (Unmethylated)

0.5 1 0.5 1

cg18384097

0.2 0.4 0.2 0.4

cg00481951

0.1 0.2 0.3 0.1 0.2 0.3

cg19671120

0.1 0.2 0.3 0.1 0.2 0.3

cg14361627

0.5 1 0.5 1

cg08928145

0.3 0.6 0.3 0.6

cg12757011

0.3 0.6 0.3 0.6

cg07547549 Methylation SNaPshot MPS Target ID Coefficient (intercept)

  • 24.521

cg18384097

  • 31.111

cg00481951 6.718 cg19671120 23.760 cg14361627 81.053 cg08928145 24.325 cg12757011 53.634 cg07547549 89.415

Model – MPS

  • 20

20 40 60 80 20 40 60 80

Predicted Age (years) Chronological Age (years)

SNaPshot model (N=95)

SNaPshot model

Target ID Coefficient (intercept)

  • 8.282

cg18384097

  • 20.730

cg00481951 126.188 cg19671120 77.801 cg14361627 121.858 cg08928145 20.599 cg12757011 1.820 cg07547549 78.596

MAD = 22.43 RMSE = 24.13

  • 20

20 40 60 80 20 40 60 80

Predicted Age (years) Chronological Age (years)

MPS vs SNaPshot model(N=95)

MPS model SNaPshot model

MAD = 3.59 RMSE = 4.72

Newly trained model

slide-9
SLIDE 9

10/10/2018 9

Further analysis

  • Various modeling

‒ Multivariate stepwise linear regression ‒ Random forest modeling ‒ Other modeling

  • Analysis tool

‒ STRait razor v3.0 ‒ Public available tools

Methylation SNaPshot vs MPS

Methylation SNaPshot MPS

Multiplex Multiplex CE based MPS / NGS (different platform) Intuitive data processing Burdensome data processing Target CpGs only Neighboring CpGs Qualitative (on-off signal) Quantitative (dye intensity) Quantitative analysis In-depth analysis A G T C

Read Sequence Index Sequence

slide-10
SLIDE 10

10/10/2018 10

Conclusion

  • Markers can

be applied to both Multiplex methylation SNaPshot and MPS.

  • The model should be altered as the platform differs.
  • Models can be varied because of more information from MPS.

Acknowledgement

Yonsei DNA Profiling Group This research was supported by the Bio & Medical Technology Development Program of the National Research Foundation of Korea (NRF) funded by the Korean government (NRF- 2014M3A9E1069992).