![1HXUREODVWRPD WXPRUV
3URPHWKD]LQH ,PLSUDPLQH
+0
+0
Approved compound for non-cancer indication prevents
formation of SCLC tumors in a genetic model of SCLC
31(7
7
10
Days of Treatment
13
5
$
3
`7 0,1
31(7V
31(7V
%
0
1
RQWURO
,PLSUDPLQH +0
,PLSUDPLQH +0 0,1
`7
,PLSUDPLQH +0
3'
0
$
0
3'
2
Mice dosed
after
tumor
formation
`7 0,1
6XUYLYDO 077](https://image.slidesharecdn.com/movingfrombigdatatobettermodelsofdiseaseanddrugresponse-joeldudley-131229113217-phpapp01/85/Moving-from-Big-Data-to-Better-Models-of-Disease-and-Drug-Response-Joel-Dudley-19-320.jpg)
![3URPHWKD]LQH ,PLSUDPLQH
+0
+0
9HKLFOH
:DWHU](https://image.slidesharecdn.com/movingfrombigdatatobettermodelsofdiseaseanddrugresponse-joeldudley-131229113217-phpapp01/85/Moving-from-Big-Data-to-Better-Models-of-Disease-and-Drug-Response-Joel-Dudley-22-320.jpg)
Moving from Big Data to Better Models of Disease and Drug Response - Joel Dudley
- 1. Moving from Big Data to Better Models of Disease and Drug Response Joel Dudley, PhD Director of Biomedical Informatics & Assistant Professor of Genetics and Genomic Sciences, Mount Sinai School of Medicine Icahn School of Medicine at Mount Sinai @IcahnIns(tute
- 2. Mount Sinai Health System Facts 7 Member hospital campuses >3,500 Hospital beds >3,100,000 Patient visits >6,000 Physicians
- 4. Mount Sinai is attracting key talent to thrive in a Big Data world Demeter
- 5. There are rarely smoking guns in human disease biology
- 6. There are rarely smoking guns in human disease biology
- 7. We must embrace complexity to fully That promise to enable the construction of molecular and that define understand human physiologynetworks disease the biological processes that comprise living systems ENVIRONMENT Non-coding RNA network ENVIRONMENT HEART protein network GI TRACT KIDNEY metabolite network IMMUNE SYSTEM VASCULATURE transcriptional network ENVIRONMENT ENVIRONMENT BRAIN
- 8. We must embrace complexity to fully understand human physiology and disease “A complex adaptive system has three characteristics. The first is that the system consists of a number of heterogeneous agents, and each of those agents makes decisions about how to behave. The most important dimension here is that those decisions will evolve over time. The second characteristic is that the agents interact with one another. That interaction leads to the third— something that scientists call emergence: In a very real way, the whole becomes greater than the sum of the parts. The key issue is that you can’t really understand the whole system by simply looking at its individual parts” . - Michael J. Mauboussin (investment banker)
- 9. Although our ability to embrace complexity will bump up against our want to tell stories Zeus, the sky god; when he is angry he throws lightening bolts out of the sky Ptolemaic astronomy: the earth is the center of the universe The earth is flat Biological processes are driven by simple linearly ordered pathways (e.g. TGF-beta signaling)
- 10. Integrating and modeling the digital universe of information
- 11. We need to be able to leverage the digital universe of information to solve complex problems 1.8 ZETTABYTES (1.8 trillion gigabytes) of information will be created and replicated in 2011and growing fast (it has grown by a factor of 9 in just five years) Last 2011 IDC Digital Ucrackedsponsored bzettabyte year WE niverse Study the 1 y EMC
- 12. Being masters of really big data is now critical for biomedical research (TB→PB→EB→ZB) Organisms Tissues Single cells Single cell, real-‐2me, con2nuous?
- 13. Real time observation systems add complex but powerful new dimensions to NGS Inter Pulse Distance (IPD)
- 14. We measure more than we know
- 15. Exploring the transcriptional landscape of human disease 20k+ Genes ~300 Diseases and Condi2ons Blue: gene goes down in disease Yellow: gene goes up in disease
- 16. Building molecular taxonomies of human disease Figure 2. Significant disease-disease similarities. (A) Hierarchical clustering of the disease correlations. The distance between two diseases wa Suthram S, Dudley J et al. Network-based elucidation of human disease similarities reveals common defined to be (1-correlation coefficient) of the two diseases. The tree was constructed using the average method of hierarchical clustering. The re functional a p-value of 0.01 and for pluripotent disease correlations below this line are considered (2010) line corresponds to modules enrichedFDR of 10.37% and, drug targets. PLoS Computational Biology significant. The different color represent the various categories of significant disease correlations. (B) The network of all the 138 significant disease correlations. The colo
- 17. Data Driven Approach to Connect Drugs and Disease Using Molecular Profiles Sirota, M., Dudley, J. T., et al. (2011). Discovery and Preclinical Validation of Drug Indications Using Compendia of Public Gene Expression Data. Science Translational Medicine, 3(96).
- 18. Topiramate Reduces IBD Severity in a TNBS Rodent Model of IBD • TNBS chemically induced rat model of IBD • Animals treated with 80mg/kg topiramate oral after sensitization • Prednisolone positive control (approved for IBD in humans) Dudley, J. T., Sirota, M., et al. (2011). Computational Repositioning of the Anticonvulsant Topiramate for Inflammatory Bowel Disease. Science Translational Medicine, 3(96).
- 19. 1HXUREODVWRPD WXPRUV 3URPHWKD]LQH ,PLSUDPLQH +0 +0 Approved compound for non-cancer indication prevents formation of SCLC tumors in a genetic model of SCLC 31(7 7 10 Days of Treatment 13 5 $ 3 `7 0,1 31(7V 31(7V % 0 1 RQWURO ,PLSUDPLQH +0 ,PLSUDPLQH +0 0,1 `7 ,PLSUDPLQH +0 3' 0 $ 0 3' 2 Mice dosed after tumor formation `7 0,1 6XUYLYDO 077
- 20. * * * * * 4 5 ** ** ** $ * ** * G 3' 6 Bepridil $ p53/Rb/p130 triple knockout model of SCLC Imipramine Promethazine F 0 ** ** * 0 Imipramine % Saline 8 1% $ c 6XUYLYDO 077
- 21. Fold Change of Tumor Volume b 1 Control F 6XUYLYDO 077
- 24. Molecular networks act as sensors and mediators of complex and adaptive cellular physiology
- 25. What we are about: Integrating big data across many domains to build predictive models that improve how we diagnose and treat disease Population Predictive Network Model Sample acquisition Slide courtesy of Eric Schadt
- 27. Causal network models generate testable predictions from in silico experiments Ultimately want to drive decision making in drug discovery Novel phosphatase under development at Merck for T2D Grit Sh3gl2 Prr7 PPM1L C6 Insulin Fat Mass Irx3 Glra2 Atp1a3 Slc38a1 Glucose Tcf7l2 Predictions derived from the predictive models Slide courtesy of Eric Schadt Increases fat mass Negatively impacts Hypertension genes Lowers glucose BAD GOOD Raises insulin
- 28. Predictions are great, but only meaningful if they are validated GLUCOSE LOWERED GOOD Grit Sh3gl2 Prr7 PPM1L C6 Insulin Fat Mass Irx3 Glra2 Atp1a3 BAD Slc38a1 Glucose FAT MASS INCREASED Tcf7l2 BLOOD PRESSURE INCREASED BAD Slide courtesy of Eric Schadt
- 29. Validation of network model But wait, the network also shows PPM1L and PPARG prediction in a patient population (target of Avandia) in a causal relationship PPARG PPM1L Network Predicts: - Avandia will lower glucose - Avandia will make you fat - Avandia will increase cardiovascular risk Validation 2 years later:
- 30. Leveraging NGS and Predictive Network Models to Drive Personalized Cancer Therapy Humancell systemscreening Pa1ent2specific mutantflymodels Pa1ent2specific xenogra7models Soma4c varia4on Clinical' CD8'epitope' predic4on Tumor'RNA' Network' integra4on Tumor'DNA' Germline'DNA' Chemo' genomic Public'data' integra4on Cancer'Pa)ent' Profiling Pa)ent0Specific' Analyses Interpreta)on''Screens' Informed'by'Pa)ent0 Specific'Tumor'Network Personalized'Report'' Treatment'Op)ons' Delivered'to'Clinician
- 31. Personalized multiscale tumor networks to diagnose and treat cancers Tumor$biopsy$+$normal Genomics Core Facility (Illumina, PacBio, Ion) RNA$+$DNA = key driver Key driver targeted therapy Patient-specific subnetwork Predictive network model of cancer
- 32. Personalized multiscale tumor networks to diagnose and treat cancers Tumor$biopsy$+$normal Genomics Core Facility (Illumina, PacBio, Ion) RNA$+$DNA = key driver Pa2ent network targeted therapy Patient-specific subnetwork Predictive network model of cancer
- 33. Personalized multiscale networks to model dynamics of complex disease DNA Cell'specific-RNA Cytokines Clinical-labs Physiometrics 0: min 00 Th1 Th17 0:05 min 0:10 min
- 34. How to capture all of the clinical data exhaust? CPOE EMR Billing Telemetry
- 35. Data driven translational medicine pipeline at Mount Sinai BioBank Research.and. Clinical.Queries; Experiment. CreaAon;.etc. PaAent. Traffic Sequencing. Facility Clinical.Labs Clinical.Data AcAonable. Feedback EMR (EPIC) Data. Warehouse Disease.Model. ConstrucAon.and. PredicAon. GeneraAon Primary.Data HighF Performance. CompuAng
- 36. Multiscale analysis of patient networks enables precision medicine = Genomic Environment Clinical
- 37. Multiscale measures of patients becoming available through the Mount Sinai Biobank Diagnoses DNA RNA Drugs Microbiome Immune Labs Procedures
- 38. Image credit: Li Li (ISMMS)
- 39. Many possible topological analyses can be driven using Mt. Sinai genotype/phenotype data Topological network generated using SNP data separates race Low enr ich Hig . di he abe tes nric h. d iab ete s DMSEA DMSAA DMSHA DMSHA, diabetes enriched
- 40. The personal biosensor wave is forming
- 43. Key challenge: incorporate data-driven models into clinical decision support at the point-of-care PRAC TICE CLIPMERGE platform Rules for actionable gene/drug pairs CRAE Genome-informed CDS Electronic health record CLIPMERGE database This patient has been prescribed clopidogrel (Plavix®) and is a CYP2C19-poor metabolizer (*2/*2) according to genomic testing. Poor metabolizer status is associated with significantly diminished antiplatelet response to clopidogrel and increased risk for adverse cardiovascular events following percutaneous coronary intervention (PCI). If no contraindication, consider alternative medication from order set below. Click here to learn more. Longitudinal clinical data Clinical genotype data Mount Sinai Genetic Testing Laboratory OK Reference material If no contraindication, consider prescribing an alternative medication. Click the medication name for further information including indications, dosage and contraindications. ® PRASUGREL (Effient ) ® TICAGRELOR (Brilinta ) OK CLIPMERGE PGx saliva sample from consented BIOMe participant Drug information Figure 1 A platform for the implementation of genome-informed clinical decision support (CDS). Saliva samples from BioMe patients sent to the Mount Sinai Genetic Testing Laboratory are subjected to clinical pharmacogenomic testing. Valid genotypes are released to the CLIPMERGE database, which also contains longitudinal clinical data extracted from the electronic health record (EHR). These data are assessed by the clinical risk assessment engine (CRAE), which contains prespecified rules relating actionable genotype–drug pairs to genome-informed advice messages. If a rule is fulfilled, decision support is delivered in real time via the EHR. A mockup of CDS for a clopidogrel (Plavix) poor metabolizer is shown, consisting of a text segment, a reference link, and an order set with suggested alternative medications. Erwin Bottinger useful genomic information, regardless of how it is generated. Omri Gottesman DEVELOPMENT AND EVALUATION OF CDS CONTENT
- 44. New from Oxford University Press • • PERSONAL GENOMICS Disease risk modeling • EXPLORING Visualization Pharmacogenomics • DNA-to-physiology • Gene-by-environment • More! JOEL T. DUDLEY KONRAD J. KARCZEWSKI Foreword by George M. Church Foreword by George Church http://exploringpersonalgenomics.org
- 45. Thank you for your attention Email: joel.dudley@mssm.edu Twitter: @jdudley Web: research.mssm.edu/dudley/ Icahn School of Medicine at Mount Sinai