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

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 Department of Forensic Medicine, Yonsei University College of Medicine, Seoul, Korea 2 Department of Forensic Medicine and Brain Korea 21 PLUS Project for Medical Science, Yonsei University, Seoul, Korea DNA Methylation • Addition of a methyl group to cytosine followed by guanine • 5’-CG-3’ 1

10/10/2018 DNA Methylation Cell differentiation Aging Aging [ ] [ ] [ ] Body Fluid Age Age Identification Prediction Prediction Genetic factor Environmental factor [ ] [ ] Genetic Behavior Traits Habits G Method • Saliva samples ‒ 280 samples (18-73 years) • HumanMethylation450 BeadChip Array ‒ 54 males (18-73 years) ‒ Marker candidates selection by multivariate linear regression analysis • Targeted Bisulfite Sequencing Info Training Set Testing Set Total Male 47 70 117 Female 48 61 109 Total 95 131 226 ‒ Multiplex methylation SNaPshot (226 samples; Both sets) ‒ Massively parallel sequencing (95 samples; Training set) ‒ Analysis using several tools (SPSS, etc.) 2

10/10/2018 Marker Selection HumanMethylation450 BeadChip Array • 6 age-associated CpG candidates + Cell type-specific marker (cg18384097) Hong et al . FSI Genet. (2017) Cell Type-specific Marker • Buccal-Cell-Signature (ϐ) 0.7 ‒ Eipel et al. Aging (Albany NY). (2016) ‒ cg07380416 ( CD6 ) 0.6 ‒ cg20837735 ( SERPINB5 ) 0.5 ‒ Percentage of buccal epithelial cells Methylation at cg18384097 R² = 0.9286 99.8 × � ���������� + 1.92 0.4 ϐ = 2 −98.12 × � ���������� + 88.54 0.3 + 2 0.2 • cg18384097 ( PTPN7 ) N = 54 0.1 ‒ Souren et al. Genome Biol. (2013) Spearman’s rho = 0.955 ‒ High in buccal epithelial cell 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 ‒ Low in blood cell Buccal-Cell-Signature (ϐ) ‒ PTPN7 gene (predicted epithelial cell compositions) - Protein tyrosine phosphatase (PTP) - Preferentially expressed in hematopoietic cells 3

10/10/2018 Detail Workflow Hong et al . FSI Genet. (2017) A G DNA Methylation Analysis Multiplex PCR Multiplex SBE 10ng Multiplex Methylation SNaPshot (N=226=95+131) Bisulfite conversed Massively Parallel Sequencing (N=95) DNA Multiplex PCR Indexing PCR Analysis tools Bismark Read Sequence Index Sequence Lee et al . FSI Genet. (2016) Methylation SNaPshot (N=95) cg00481951 (SST) cg19671120 (CNGA3) cg14361627 (KLF14) 0.4 0.4 0.4 R² = 0.5412 R² = 0.2541 R² = 0.6313 0.3 0.3 0.3 Methylation Methylation Methylation 0.2 0.2 0.2 0.1 0.1 0.1 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) cg07547549 (SLC12A5) cg12757011 (TBR1) cg08928145 (TSSK6) 1 0.6 0.6 R² = 0.5928 R² = 0.4193 R² = 0.2139 0.5 0.5 0.8 Methylation Methylation Methylation 0.4 0.4 0.6 0.3 0.3 0.4 0.2 0.2 0.2 0.1 0.1 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) 4

10/10/2018 Methylation SNaPshot (N=226) cg00481951 (SST) cg19671120 (CNGA3) cg14361627 (KLF14) 0.4 0.4 0.4 R² = 0.6347 R² = 0.4791 R² = 0.2882 0.3 0.3 0.3 Methylation Methylation Methylation 0.2 0.2 0.2 0.1 0.1 0.1 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) cg07547549 (SLC12A5) cg12757011 (TBR1) cg08928145 (TSSK6) 1 0.6 0.6 R² = 0.4341 R² = 0.167 R² = 0.5486 0.5 0.5 0.8 Methylation Methylation Methylation 0.4 0.4 0.6 0.3 0.3 0.4 0.2 0.2 0.2 0.1 0.1 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) Model – Multiplex Methylation SNaPshot Training Set (N=95) Testing Set (N=131) Target ID Coefficient 80 80 (intercept) -24.521 -31.111 cg18384097 Predicted Age (years) 60 Predicted Age (years) 60 cg00481951 6.718 23.760 40 40 cg19671120 cg14361627 81.053 20 20 cg08928145 24.325 MAD = 3.03 MAD = 3.43 RMSE = 4.03 RMSE = 4.36 53.634 cg12757011 0 0 0 20 40 60 80 0 20 40 60 80 cg07547549 89.415 Chronological Age (years) Chronological Age (years) MAD: Mean Absolute Deviation RMSE: Root Mean Square Error 5

10/10/2018 MPS Analysis • Platform ‒ MiSeq Reagent Kit v3 (2×300) ‒ HiSeq 2000 • Pipeline Integrated Raw CpG Trimming Bismark data file report ‒ Adapter ‒ Alignment ‒ Quality ‒ CpG Extraction MPS (N=95) – CpG sites in amplicons • Pearson’s R (Correlation between chonological age and methylation) CpG CpG CpG CpG CpG CpG CpG CpG CpG CpG CpG CpG CpG CpG CpG ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 * Statistically significant 6

10/10/2018 MPS (N=95) – Methylation value cg00481961_CpG3 cg19671120_CpG15 cg14361627_CpG6 0.2 0.25 0.2 R² = 0.6626 R² = 0.5714 R² = 0.3219 0.2 Methylation Methylation Methylation 0.15 0.1 0.1 0.1 0.05 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) cg07547549_CpG9 cg08928145_CpG13 cg12757011_CpG4 1 0.5 0.4 R² = 0.4043 R² = 0.5917 R² = 0.2393 Methylation Methylation Methylation 0.5 0.25 0.2 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) Age Prediction Using the MPS Data SNaPshot model (N=95) 80 Target ID Coefficient (intercept) -24.521 60 cg18384097 -31.111 Predicted Age (years) cg00481951 6.718 40 cg19671120 23.760 SNaPshot model 20 cg14361627 81.053 MAD = 22.43 cg08928145 24.325 RMSE = 24.13 0 cg12757011 53.634 0 20 40 60 80 cg07547549 89.415 -20 Chronological Age (years) 7

10/10/2018 Methylation SNaPshot vs MPS (N=95) cg18384097 cg00481951 cg19671120 cg14361627 1 0.3 0.4 0.3 0.2 0.2 0.5 0.2 0.1 0.1 0 0 0 0 0 0.5 1 0 0.2 0.4 0 0.1 0.2 0.3 0 0.1 0.2 0.3 cg08928145 cg12757011 cg07547549 1 0.6 0.6 0.5 0.3 0.3 G (Methylated) MPS A 0 0 0 (Unmethylated) 0 0.5 1 0 0.3 0.6 0 0.3 0.6 Methylation SNaPshot Model – MPS Newly trained model MPS vs SNaPshot model(N=95) SNaPshot model (N=95) 80 80 Target ID Target ID Coefficient Coefficient (intercept) (intercept) -24.521 -8.282 60 60 cg18384097 cg18384097 -20.730 -31.111 Predicted Age (years) Predicted Age (years) MAD = 3.59 RMSE = 4.72 cg00481951 cg00481951 126.188 6.718 40 40 MPS model cg19671120 cg19671120 23.760 77.801 SNaPshot model SNaPshot model 20 20 cg14361627 cg14361627 121.858 81.053 MAD = 22.43 cg08928145 cg08928145 20.599 24.325 RMSE = 24.13 0 0 cg12757011 cg12757011 53.634 1.820 0 0 20 20 40 40 60 60 80 80 cg07547549 cg07547549 78.596 89.415 -20 -20 Chronological Age (years) Chronological Age (years) 8

10/10/2018 Further analysis • Analysis tool ‒ STRait razor v3.0 ‒ Public available tools • Various modeling ‒ Multivariate stepwise linear regression ‒ Random forest modeling ‒ Other modeling Methylation SNaPshot vs MPS Methylation SNaPshot MPS G A Read Sequence C T Index Sequence Multiplex Multiplex CE based MPS / NGS (different platform) Intuitive data processing Burdensome data processing Target CpGs only Neighboring CpGs Qualitative (on-off signal) Quantitative analysis Quantitative (dye intensity) In-depth analysis 9

10/10/2018 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 ). 10