The Foundation of P4 Medicine Keynote Presentation as presented by Leroy Hood, M.D., PhD, at the Ohio State University Personalized Health Care National Conference 2010.

1. The Foundations of P4 MedicinePredictive, Personalized, Preventive and Participatory Lee Hood Institute for Systems Biology, Seattle

2. Outline Principles of P4 medicine P4 medicine— and the future Societal implications of P4 medicine ISB strategic partnerships for P4 medicine

3. The Foundation of P4 Medicine – Four Concepts View medicine as an informational science Systems approaches allow one to understand wellness and disease—holist rather than atomistic Emerging technologies will allow us to explore new dimensions of patient data space Transforming analytic tools will allow us to decipher the billions of data points for the individual--sculpting in exquisite detail wellness and disease

4. The Foundation of P4 Medicine – Four Concepts View medicine as an informational science Systems approaches allow one to understand wellness and disease—holist rather than atomistic Emerging technologies will allow us to explore new dimensions of patient data space Transforming analytic tools will allow us to decipher the billions of data points for the individual--sculpting in exquisite detail wellness and disease

6. All Hierarchical or MultiscaleLevels of Biological Information—Are Modified by Environmental Signals DNA RNA Protein Protein interactions and biomodules Protein and gene networks Cells Organs Individuals Populations Ecologies

7. The Foundation of P4 Medicine – Four Concepts View medicine as an informational science Systems approaches allow one to understand wellness and disease—holist rather than atomistic (systems biology and systems medicine Emerging technologies will allow us to explore new dimensions of patient data space Transforming analytic tools will allow us to decipher the billions of data points for the individual--sculpting in exquisite detail wellness and disease

8. ISB’s View of Systems Biology

9. Essentials of Systems Biology Hypothesis-driven and hypothesis-generating Data Global data acquisition Integrate multi scale data types Delineate biological network dynamics—temporal and spatial Dealing with biological and technical noise in large data sets Formulate models that are predictive and actionable. Discovery science is key

11. Chemistry

12. Computer Science

13. Engineering

14. Mathematics

15. PhysicsTECHNOLOGY COMPUTATION

16. Institute for Systems Biology Founded 2000—10th Anniversary ISB has 13 faculty and 300 staff

17. ISB’s description of systems biology in 2000 is virtually identical to that of this National Academy of Sciences 2010 report entitled the “New Biology”. ISB was the first Systems Biology organization in 2000—today there are more than 70 world wide Predicted that systems approaches would drive medicine of the future

18. SCImago Institutions Rankings: http://www.scimagoir.com/ Inst. Nat. de Physique Nucleaire Sanger SSM Cardinal Glennon Children’s Hospital ISB CSHL International Agency for Research on Cancer Kaiser Permanente Brigham and Women’s Hospital Rockefeller Institut Catala d’Investigacio Quimica MIT Salk WHO Harvard HHMI MSFT FHCRC IBM Parc Mediterrani de la Tecnología Centro de Investigación Príncipe Felipe CNRS Chinese Academy of Science Russian Academy of Sciences National Academy of Sciences of Ukraine Tianjin University Harbin Engineering University Research Institute of Petroleum Processing ISB 1st in US and 3rd in World for Impact of Papers

19. A Systems View of Disease

20. dynamics of pathophysiology diagnosis therapy prevention A Systems View of Medicine Postulates that Disease Arises from Disease-Perturbed Networks Non-Diseased Diseased

21. Neuropathology Identifies 4 Aspects of PrionNeurodegeneration Microglia / Astrocyte activation PrP accumulation Synaptic Degeneration Nerve cell death Infected Normal

22. Dynamics of a Brain Network in Prion Neurodegenerative Disease in Mice Prion accumulation network

23. Sequential Disease-Perturbation of the Four Networks of Prion Disease 18~20 wk 22 wk 0 wk Clinical Signs Prion accumulation Glial Activation SynapticDegeneration Neuronal Cell Death Na+ channels Reactive Astrocytes Cholesterol transport Caspases Sphingolipid synthesis Cargo transport Leukocyte extravasation Lysosome proteolysis *Arachidonate metab./Ca+ sig.

24. DEGs Encoding Known and Novel Prion Disease Phenotypes 333 DEGs encode core prion disease 231/333 DEGs encode known 4 disease-perturbed networks from histopathology 102/333 DEGs encode 6 novel disease-perturbed networks--the dark genes of prion disease Disease-perturbed networks sequentially activated Re-engineer disease-perturbed networks with drugs—new approach to drug target discovery Striking implications for blood diagnostics

25. Dynamics of a Brain Network in PrionNeuroddegenerative Disease in Mice Prion accumulation network

26. Making Blood A Window Distinguishing Health and DiseaseOrgan-specific Blood Proteins 110 brain-specific blood proteins/80 liver-specific blood proteins Blood Vessel

27. Why Systems-Driven Blood Diagnostics Will Be the Key to P4 Medicine Early detection Disease stratification Disease progression Assess prognosis Follow therapy Assess reoccurances

28. The Foundation of P4 Medicine – Four Concepts View medicine as an informational science Systems approaches allow one to understand wellness and disease—holist rather than atomistic Emerging technologies will allow us to explore new dimensions of patient data space Transforming analytic tools will allow us to decipher the billions of data points for the individual--sculpting in exquisite detail wellness and disease

29. ISB’s Technology-Driven Big Biology Projects

30. Four ISB Technology-Driven New Big Projects The Human Proteome Project—SRM mass spectrometry assays for all human proteins Clinical assays for patients that allow new dimensions of data space to be explored Complete genome sequencing of families—sequences and new stratifications to identify disease genes—1000s individuals The 2nd Human Genome Project—mining all complete human genomes and their phenotypic/clinical data

31. Whole Genome Sequencing of Families: New Genomic Strategy Sequencing by Complete Genomics, Inc.

32. Whole Genome Sequencing of Family of Four Unaffected parents Children each with 2 diseases--craniofacial malformation (Miller Syndrome) and lung disease (ciliarydyskinesia) Identify 70% of sequence errors using principles of Mendelian genetics —less than 1/100,000 error rate Discovery of about 230,000 rare variants in family—confirmed by identification in two or more family members

33. Recominational Genome Map from Miller’s Syndrome Children: 65% of recombinational events fall in hot spots Both children inherited the same allele from both parents Each child inherited a different allele from each parent Children inherited the same allele from dad, different alleles from mom Children inherited the same allele from mom, different alleles from dad x x x 65 crossovers in (2) male meioses (left) 104 crossovers in (2) female meioses (right) 250 200 More than 65% crossovers In hot spots of recombination 150 100 50 0 X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

35. New mutations, germline nucleotide substitutions, have never been directly measured before.

36. Low error rate & family makes this approach work

37. We trapped by Agilent hybridization selection and resequenced ~60,000 sites

38. Indirect, phylogenetic estimate is between 8 x 10-9 &2.1 x 10-8 per base per generation,

39. The intergenerational mutation rate of the autosome is

41. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X ZNF721 DNAH5 KIAA0556 DHODH Miller’s gene Ciliarydyskenesis gene centromere paternal recombination heterochromatin identical error region maternal recombination CNV haploidentical paternal candidate gene

42. Family Genome Sequencing May Facilitate Finding Mendeliandisease genes Modifiers of disease genes--sequencing genomes of 65 Huntington’s patients from families Genes encoding complex genetic diseases after proper patient stratification—Alzheimer’s/Parkinson’s diseases

43. The Human Proteome Project Strategic partners: ISB (R. Moritz)/ETH (R. Aebersold)/Agilent/AB-Sciex/Origene

44. ISB has made 4 fundamental contributions to the human proteome project

45. 1. Trans Proteomic Pipeline (TPP) components * Commercial software not part of TPP

46. Drives tool development and optimization Advanced, uniform processing of all data 2. TPP: Foundation for PeptideAtlas

47. 3. Targeted Proteomics: Human SRMAtlas SRM assays for most of the known 20,333 human proteins 39

48. ISB Contributions to the Human Proteome Project Trans Proteomic Pipeline—quality assessment/validation Protein Atlas—large-scale MS database Pioneered SRM/MRM mass spectrometry applications—a targeted approach to protein assays 4. First preliminary SRM assays (map) of all 20,333 human proteins

49. Proposal for a Human Proteome Project Create validated targeted assays (SRM) for each human protein Do the same for carefully selected model organisms (mouse, rat, fly, nematode, etc.) Develop diverse technologies to increase the power of proteome analyses—MS, protein chips, novel protein capture agents, imaging, single protein molecule analyses, etc. Develop the computational and mathematical tools necessary for proteome analyses Develop the software to make all proteins in their biologically relevant clusters--biological networks/molecular machines--accessible to all biologists With appropriate time lines build in specific aims to include the many other dimensions of proteomics (7-10 yr project)

50. Microfluidic Protein Chip:Assay 2500 Organ-Specific Blood Proteinsfrom Millions of Patients Using just a Drop of Blood—Follow Health Longitudinally and Detect Transitions to Disease Jim Heath--Caltech

51. DEAL for In vitro molecular diagnostics: Integrated nanotech/microfluidics platform 300 nanoliters of plasma cells out Assay region 5 minute measurement Jim Heath, et al

52. Peptide Protein-Capture AgentsJim Heath--Caltech Jim Heath--Caltech

53. Antibody Displacement Technology—Heath--Caltech An example of a triligand PCC agent for bovine carbonic anhydrase II Protein Catalyzed Capture Agents: triligands determined by repeated screening of target protein across synthetic bead-bound peptide libraries anchor peptide is selected on the first screen protein catalyzes the formation of second ligand to anchor ligand on second screen protein catalyzes the formation of the third ligand to the anchor and second ligand on third screen high affinity, stable and easily manufactured triligand capture agents confidential 45 Jim Health et. al.,

54. Single-Cell Analysis

55. Quantitative transcriptome clustering of single cells from the human glioblastoma cell line U87 CD markers let us sort into 3 quantized cell populations

56. Technologies for Exploring New Dimensions of Patient Data Space

57. Individual Patient Information-Based Assays of the Present/ Future (I) Genomics Complete individual genome sequences—predictive health history—will be done sequencing families Complete individual cell genome sequences—cancer. Complete MHC chromosomal sequence in families—autoimmune disease and allegies 200 Actionable SNPs—pharmacogenetics-related and disease-related genes Sequence 1000 transcriptomes—tissues and single cells—stratification disease Analyze aging transcriptome profiles—tissues and single cells—wellness Analyze miRNA profiles—tissues, single cells and blood—disease diagnosis Proteomics Organ-specific blood MRM protein assays—110 brain, 80 liver and 20 lung 2500 blood organ-specific blood proteins from 300 nanoliters of blood in 5 minutes—twice per year (50 proteins from 50 organs)—wellness assessment. New protein capture agents. Array of 13,000 human proteins—against autoimmune or allergic sera--stratify. Single molecule protein analyses—blood organ-specific proteins and single cell analyses

58. Individual Patient Information-Based Assays of the Present/ Future (II) Single cells Analyze10,000 B cells and 10,000 T cells for the functional regions of their immune receptors—past and present immune responsiveness—follow vaccinations—identify autoimmune antibodies. Analyze individual blood macrophages—inflammation, etc. Use pore technology to separate epithelial cells from blood cells—cancer iPS (stem) cells Analyze individual stem (iPS) cells from each individual differentiated to relevant tissues to get important phenotypic information—molecular, imaging and higher level phenotypic measurements.

59. Induced Pluripotent Stem Cells (iPS Cells)

60. Stratification of Complex Genetic Diseases—e.g.Alzheimer’s Disease Collect families of patients with the relevant disease (families will stratify disease to certain extent) Create iPScells from each individual in families Differentiate iPS cells to neuronsin test tubes Probeindividual neurons with single cell transcriptomeanalyses to identify the degree of heterogeneity and neuron types—quantification of cell types--cell sorting if necessary Probethese neurons (individually or as cell-sorted classes) with ligands, drugs and relevant RNAi’s and analyzetheir transcriptome, miRNAome and selected proteomes—this will stratify different combinations of disease-perturbed (or potentially disease-perturbed) networks Sequence family genomes in keeping with their initial stratification types for error corrections Global comparisons of data across and within families of the molecular data—for final disease stratification

61. The Foundation of P4 Medicine – Four Concepts View medicine as an informational science Systems approaches allow one to understand wellness and disease—holist rather than atomistic Emerging technologies will allow us to explore new dimensions of patient data space Transforming analytic tools will allow us to decipher the billions of data points for the individual sculpting in exquisite detail wellness and disease

62. Phenome Transcriptome Transactional Epigenome Single Cell iPS Cells Social Media TeleHealth UUAGUG AUGCGUCUAGGCAUGCAUGCC Na143 K 3.7 BP 110/70 HCT32 BUN 12.9 Pulse 110 PLT150 WBC 92 110101000101010101101010101001000101101010001 110101000101010101101010101001000101101010001 110101000101010101101010101001000101101010001 110101000101010101101010101001000101101010001 110101000101010101101010101001000101101010001 110101000101010101101010101001000101101010001 Genome GCGTAG ATGCGTAGGCATGCATGCCATTATAGCTTCCA Proteome arg-his-pro-gly-leu-ser-thr-ala-trp-tyr-val-met-phe-asp-cys Billions of Data Points Are Emerging Around Each Individual

63. The Foundation of P4 Medicine – Four Concepts View medicine as an informational science Systems approaches allow one to understand wellness and disease—holist rather than atomistic Emerging technologies will allow us to explore new dimensions of patient data space Transforming analytic tools will allow us to decipher the billions of data points for the individual--sculpting in exquisite detail wellness and disease

64. Predictive, Personalized, Preventive and Participatory (P4) Medicine Driven by systems approaches to disease, new measurement (nanotechnology) and visualization technologies and powerful new computational tools, P4 medicine will emerge over the next 10-20 years 56

65. P4 medicine—the future

67. Probabilistic health history--DNA sequence

68. Biannual multi-parameter blood protein measurements

70. Unique individual human genetic variation mandates individual treatment

71. Patient is his or her own control—longitudinal data

72. Billions of data points on each individual

76. Patient understands and participates in medical choices

77. Physicians trained before P4 will have to understand it

78. Medical community—interconnected and educated

80. ISB’s Strategic Partners for P4 Medicine Develop the P4 tools and strategies for patient assays—State of Luxembourg--$100 million over 5 years Bring P4 medicine to patients with the creation of the non-profit P4 Medical Institute (P4MI) in partnership with Ohio State Medical School

81. The P4 Medicine Institute (http://www.P4MI.org) Non-profit 501c3--ISB and Ohio State founding members Vision--identify, recruit and integrate strategic partners to bring P4 medicine to patients Convince a skeptical medical community through promotion two Ohio pilot projects (and others)—wellness and lung cancer—exhibit success and power of P4 medicine Seek academic and industrial partners who share the P4 vision and have complementary skills/resources Bringing on consultants to analyze the societal challenges of P4 medicine—ethics, security, confidentiality, policy, regulation, economics, etc.

82. 9 Dimensions of P4 Medicine (I) P4 medicine is medicine of the present/near future. P4 medicine is revolutionaryrather than evolutionary or incremental P4 medicine is drivenbysystems approaches to disease and emerging technologies P4 medicine will use measurements to quantify wellness and its transition into disease P4 medicine sees the patient (consumer) as the central focus of healthcare

83. 9 Dimensions of P4 Medicine (II) Pilot projects with informational assays in patient groups will be necessary to convince skeptics. P4 medicine will restructure the business plans of every sector of the healthcare industry—enormous economic opportunities P4 medicine will be effective and inexpensive—readily available to poor and rich. The national healthcare debate in the future should be reframed around P4 medicine rather than the old reactive medicine.

84. Conceptual Themes of P4 Medicine Disease Demystified Wellness Quantified P4 Medicine Predictive Preventive Personalized Participatory

85. The Grand Challenge of the 21st Century in Science and Technology Is Complexity New concepts, strategies and technologies permit biologists to successfully begin to attack biological and medical complexity View biology as an informational science Systems approaches permit one to attack complexity effectively Evolving current and emerging technologies permit the exploration of new areas of data space (and improve the old) Computation and mathematical tools permit one to acquire, store, transmit, integrate, mine and create predictive models. These approaches will allow us to effectively attack some of society’s most vexing challenges—healthcare (P4 medicine), global health, environment, energy, nutrition, agriculture, etc.

86. Acknowledgements Prion--McLaughlin Research Institute Great Falls, Montana RanjitGiri Douglas Spicer Rajeev Kumar Rose Pitstick Rebecca Young George A. Carlson Family genome project—ISB/UW/Utah/Complete Genomics—David Galas P4MI Institute—Fred Lee, Mauricio Flories, Clay Marsh (OSU) Single protein analysis—Chris Laustead Brain imaging—Nathan Price (UI)UI) Prion--Institute for Systems Biology Daehee Hwang Inyoul Lee HyuntaeYoo Eugene Yi (proteomics core facility) BruzMarzolf (Affymetrix core facility) Nanotechnology—protein chips, protein-capture agents--Jim Heath, Caltech SRM protein assays and Human Proteome—R Moritz, R Aebersold, OriGene and Agilent Single-cell analyses—Leslie Chen and QiangTian Luxemburg Strategic Partnership—David Galas, Diane Isonaka, Rudi Balling (Lux)