1. The Genomics Revolution and Human
Health
Michael Snyder
August 15, 2013
Conflicts: Personalis, Genapsys, Illumina

2. Health Is a Product of Genome +
Environment
Exposome
Health
Genome

3. Health Is a Product of Genome +
Environment
Exposome
Health
Genome

4. • Understand and Treat Disease
– Cancer
– Mystery diseases
• Pharmacogenomics
– Determining which drug side effects and doses
• Managing Health Care in Healthy Individuals
Impact of Genomics on Medicine

5. Personalized Omics Profiling: Combine
Genomic and Other Omic Information
Genomic Transcriptomic, Proteomic,
Metabolomic
1. Predict risk
2. Diagnose
3. Monitor
4. Treat &
5. Understand
Disease States
GGTTCCAAAAGTTTATTGGATGCCGTT
TCAGTACATTTATCGTTTGCTTTGGAT
GCCCTAATTAAAAGTGACCCTTTCAAA
CTGAAATTCATGATACACCAATGGATA
TCCTTAGTCGATAAAATTTGCGAGTAC
TTTCAAAGCCAAATGAAATTATCTATG
GTAGACAAAACATTGACCAATTTCATA
TCGATCCTCCTGAATTTATTGGCGTTA
GACACAGTTGGTATATTTCAAGTGACA
AGGACAATTACTTGGACCGTAATAGAT
TTTTTGAGGCTCAGCAAAAAAGAAAAT
GGAAATTAATTTTGAAGTGCCATTGA
….

6. Genome
Transcriptome
(mRNA, miRNA, isoforms, edits)
Proteome
Metabolome
Personal
Omics
Profile
Autoantibody-ome
Microbiome
Personal “Omics” Profiling (POP)
Cytokines
Epigenome

7. Genome
Transcriptome
(mRNA, miRNA, isoforms, edits)
Proteome
Metabolome
Personal
Omics
Profile
Autoantibody-ome
Microbiome
Personal “Omics” Profiling (POP)
Cytokines
Epigenome
Initially
40K
Molecules/
Measure-
ments
Now
Billions!

8. Personal Omics Profile
40 months; 61 Timepoints; 6 Viral Infections
/
/
Chen et al., Cell 2012

9. Accurate Genome Sequencing
3.3 M Hi conf. SNVs, 217K Indels and 3K SVs
2 or more Platforms
(Plus low confidence)
Whole Genome Sequencing
• Complete Genomics: 35 b paired ends (150X)
• Illumina: 100 b paired ends (120X)
Exome Sequencing
• Nimblegen
• Illumina
• Aglilent
3.30M
89%
100K
2%
345K
9%
CGIllumina

10. Local phasing + population data= highly phased blocks
Moleculo: Volodymyr Kuleshov, Michael Kertesz
Percent SNPs phased 98.2%
Switch accuracy 99.9%+

11. CodingNon-Coding
miRNA Splice UTR
miRNA
targets
Seed
sequence SIFT PP2
OMIM/Curated
Mendelian disease
(51)
Nonsynonymous
(1320)
Synonymous
mRNA
stability
tRNA
rate
I. Highly Penetrant
Variants:
Mendelian Disease
Risk Pipeline
Rick Dewey & Euan Ashley
Damaging
(234)
All variants
~3.5M
Rare/novel
variants (<5%)

12. Missense
• ALAD, ABCC2, ACADVL, ADAMTS13, AGRN, BAAT, CDS1, CHD
7, COL4A3, CTSD, DGCR2, DLD, DYSF, EPCAM, FGFR1OP, FKR
P, GAA, GNAI2, HSPB1, IGKC, ITPR1, MED12, MKS1, NTRK1, P
CM1, PKD1, PLEKHG5, PMS2, PRSS1, PTCH2, SERPINA1, SETX,
SYNE1, TERT, TTN, VWF, ZFPM2, PNPLA2.
Nonsense
• PRAMEF2, PLCXD2, NUP54, RP1L1, PIK3C2G, ND
E1, GGN, CYP2A7, IGKC
Not Rare But Important
• KCNJ11 , KLF4, GCKR …
High
Cholesterol
Aplastic
Anemia
Rare Variants in Disease Genes (51 Total)

13. Integrate Over Many Markers:
Complex Disease
0% 100%
Predict Type 2 Diabetes

14. "# )"# %""# %)"# 200 250 300 350 400 450 500 550 600 650# +%""# +)"# 0 )"# %""# %)"# &""# &)"# '""# ')"# (""# ()"# )""#
HRV Infection
(Day 0-21)
RSV Infection
(Day 289-311)
Life Style Change
(Day 380-Current)
Glucose(mg/dL)
Day Number (Relative to 1st Infection)
80
90
100
110
120
130
140
150
160
-150
Glycated HgA1c (%):
(Day Number)
6.4
(329)
6.7
(369)
4.9
(476)
5.4
(532)
5.3
(546)
4.7
(602)
GLUCOSE LEVELS
HRV INFECTION
(DAY 0-21)
RSV INFECTION
(DAY 289-311)
LIFESTYLE
CHANGE
(DAY 380-
CURRENT)
14
HbA1c (%): 6.4 6.7 4.9 5.4 5.3 4.7
(Day Number) (329) (369) (476) (532) (546) (602)

15. Dynamical Outcomes for Integrated Analysis of
Proteome, Transcriptome, Metabolome
george mias
RSV 18 days
Platelet Plug Formation
Glucose Regulation of
Insulin Secretion

16. The Future?
Genomic Sequencing
1. Predict risk
2. Early Diagnose
3. Monitor
4. Treat
Omes and Other Information:
Home Sensors
http://www.baby-connect.com/
GGTTCCAAAAGTTTATTGGATGC
CGTTTCAGTACATTTATCGTTTG
CTTTGGATGCCCTAATTAAAAGT
GACCCTTTCAAACTGAAATTCAT
GATACACCAATGGATATCCTTAG
TCGATAAAATTTGCGAGTACTTT
CAAAGCCAAATGAAATTATCTAT
GGTAGACAAAACATTGACCAATT
TCATATCGATCCTCCTGAATTTAT
TGGCGTTAGACACAGTTGGTATA
TTTA….

17. Study of 10 Healthy People
5 Asian, 5 European
Dewey, Grove, Pan, Ashley, Quertermous et al
- Median 5 reportable disease risk
associations (ACMG) per individual (range
2-6)
- 3 followup diagnostic tests (range 0-10)
- Cost $362-$1427 per individual
- 54 minutes per variant

18. Many Unaddressed Challenges
1) Accuracy and coverage
2) Interpretation
1) Interpreting non-protein
coding regions
2) DNA Methylation
5) Sample size
6) Exposome

19. 1) Accurate Genome Sequences
and Coverage
Whole Genome Sequencing
• Complete Genomics: 35 b paired ends (150X)
• Illumina: 100 b paired ends (120X)
3.30M
89%
100K
2%
345K
9%
CGIllumina
Single Nucleotide Variants Getting Better.
Indels and Structural Variants Need Work!

20. SNV Comparison
• Complete Genomics: 35 b paired ends (150X)
• Illumina: 100 b paired ends (120X)
3.30M
89%
100K
2%
345K
9%
Complete
Genomics
Illumina
Hugo Lam, Michael Clark, Rui Chen
Ti/Tv = 1.68
17/18 Sanger
Ti/Tv = 2.14
20/20 Sanger
Ti/Tv = 1.40
2/15 Sanger
31 Disease
Associated SNP
3 Disease
Associated SNP

21. Sequencing Accuracy
Sequencing the Same Genome Twice
Personalis
146,100 SNPs (3.7%)

22. Exome-seq and WGS-specific detection
45X WGS vs 80X Exome
Clark et al. 2011 Nature Biotech

23. Overall Statistics for Finishing Medically
Interesting Genes- ACE
ACE v1 = Thick Lines
TruSeq Exome (10G) = Thin
Lines
Personalis
Normal Exome ~2,000
Custom Exome (ACE) ~2,000

24. Exons Covered by ACE, Missed
by Standard Exome
Personalis

25. 1) Search for disease causing mutations
(highly penetrant)
GCKR (high lipids); TERT (aplastic anemia)
2) Sum over multiple common risk allele to
predict risk
2. Genome Interpretation
Missense Variants
ALAD, ABCC2, ACADVL, ADAMTS13, AGRN, BAAT, CDS1, CHD7, COL4A3, CTSD
, DGCR2, DLD, DYSF, EPCAM, FGFR1OP, FKRP, GAA, GNAI2, HSPB1, IGKC, ITP
R1, MED12, MKS1, NTRK1, PCM1, PKD1, PLEKHG5, PMS2, PRSS1, PTCH2, SERPI
NA1, SETX, SYNE1, TERT, TTN, VWF, ZFPM2, PNPLA2.
0% 100% Ashley, Butte et al.

26. Missing Regulatory Variation
88% of Disease Variants Lie Outside of Genes!
26
X
Two approaches:
1) Mapping transcription factor binding in different
people.
2) RegulomeDB: Assembling regulatory information from
the ENCODE Project and other sources.

27. Damaging Variation in an Individual
Gene Regulatory region
Protein Coding Non-coding
and
CAPN1: Protective against Alzheimer’s
Coding Variants
Regulatory Variants

28. 3. Incorporate Methylation Data

29. Possible Phenotypic Consequences of
Differentially Methylated Regions?

30. 4. Sample Size—Need to Reduce

31. AliveCor Measures ECG
5. Other Data Types: Sensors
71
Moves App

32. Conclusions
1) Personal genome sequencing is here. The
medical interpretation is difficult.
2) Genome sequencing can predict disease risk that
can be monitored with other omics information.
3) Integrated analysis can provide a detailed
physiological perspective for what is occurring.
4) Every person’s complex disease profile is different
and following many components longitudinally
may provide valuable information.
5) You are responsible for your own health
Data at: snyderome.stanford.edu

33. The Personal Omics Profiling Project
Rui Chen, George Mias, Hugo Lam, Jennifer Li-Pook-
Than, Lihua Jiang, Konrad Karczewski, Michael
Clark, Maeve O’Huallachain, Manoj Hariharan,Yong
Cheng, Suganthi Bali, Sara Hillemenyer, Rajini
Haraksingh, Elana Miriami, Lukas Habegger, Rong
Chen, Joel Dudley, Frederick Dewey, Shin Lin, Teri
Klein, Russ Altman, Atul Butte, Euan Ashley, Tom
Quetermous, Mark Gerstein, Kari Nadeau, Hua Tang,
Phyllis Snyder

34. Acknowledgements
34
Human Regulatory Variation:
Maya Kasowski, Fabian Grubert, Alex Urban, Alexej
A, Chris Heffelfinger, Manoj Harihanan, Akwasi
Asbere, Lukas Habegger, Joel Rozowsky, Mark
Gerstein, Sebastian Waszak, Jan Korbel
(EMBL, Heidelberg)
Regulome DB:
Alan Boyle, Manoj Hariharan, Yong Cheng, Eurie
Hong, Mike Cherry
Methylome:
Dan Xie, Volodymyr Kuleshov, Rui Chen, Dmitry
Pushkarev, Konrad Karczewski, Alan Boyle, Tim
Blauwkamp, Michael Kertesz

36. Genome (1TB)
Transcriptome (0.7TB)
(mRNA, miRNA, isoforms, edits)
Proteome (0.02 TB)
Metabolome (0.02 TB)
Personal
Omics
Profile
Total =
5.74TB/Sa
mple +
1 TB
Genome
Autoantibody-ome
Microbiome (3TB)
6. Big Data Handling and Storage
Cytokines
Epigenome (2TB)