TRANSFORMING

BIG DATA INTO

CLINICALLY ACTIONABLE

STRATEGIES

John Quackenbush

Dana-Farber Cancer Institute

Harvard School of Public Health

Background and Disclosures• Professor of Biostatistics and Computational

Biology, Dana-Farber Cancer Institute

• Professor of Computational Biology and

Bioinformatics, Harvard School of Public Health

• Many other academic titles

• Numerous advisory boards

• Co-Founder of Genospace, a Precision Genomic

Medicine Software Company (now owned

by HCA).

Background and Disclosures• Professor of Biostatistics and Computational

Biology, Dana-Farber Cancer Institute

• Professor of Computational Biology and

Bioinformatics, Harvard School of Public Health

• Many other academic titles

• Numerous advisory boards

• Co-Founder of Genospace, a Precision Genomic

Medicine Software Company (now owned

by HCA).

Every revolution in science—from

Copernican heliocentric model to the

rise of statistical and quantum

mechanics, from Darwin’s theory of

evolution and natural selection to the

theory of the gene—has been driven

by one and only one thing: access to

data.

–John Quackenbush

@johnquackenbush-Every revolution in

the history science has been driven by

one and only one thing: access to data.

Twitter version, 115 characters with spaces

@notrealdonaldtrump-We have the best

data. Fantastic Bigly! The other

scientists. losers So sad #fakedata

#failedscientists covfefe

Twitter version, 131 characters with spaces

Costs of Generating Data Have Plummeted

New Sources of Health and Medical Data

Drug Research Social Media Patient Records Genomics

Test Results Claims Data Home Monitoring Mobile Apps

http://ihealthtran.com/wordpress/2013/03/infographic-friday-the-body-as-a-source-of-big-data/

The Body as a Source of Big Data

• It is estimated that by 2015,

the average hospital will

generate 665TB of data

annually.

• Medical imaging archives are

increasing by 20-40% annually.

• Today, 80% of data is

unstructured such as images,

video, and notes.

Graphic from NetApp

Building Data-Driven

Precision Medicine

CHALLENGES AND KEY TRENDS

EXPLOSION OF GENOMIC DATA

$1,000

$10,000

$100,000

$1,000,000

$10,000,000

$100,000,000

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

COST PER GENOME

RANGE OF CLINICAL RELEVANCE

COMPUTING RESOURCES

• Data Transfer• Security/Permissioning• Storage• Compute Resources• Semantic Databases

VARIOUS DATA SOURCES IN SILOS

EMR(clinical,

outcomes)

DNA(genomic)

Other Lab

(diagnostic)

VARIOUS USE CASES AND STAKEHOLDERS

PatientClinic Lab R&D

POPULATION ANALYTICS: COMPLEX ANALYSIS MADE EASY

©Genospace, 2012-2015

COMMUNITY PORTAL: SELF-REPORTED DATA

Springfield Diagnostic Labs

CONFIDENTIAL

PRECISION MEDICINE DEMANDS SIMPLICITY

Curation, curation, curation….

The Need for

Data-Driven

Precision Medicine

Biomarkers, Disease Subtypingand Targeted Therapy

Her-2+

(Herceptin)

(Perjeta)

EML4-ALK

(Xalkori)

K-ras

(Erbitux)

(Vectibix)

BRAF-V600

(Zelboraf)

CFTR-G551

(Kalydeco)

Companion Diagnostics: the Right Rx for the Right Disease (Subtype)

Non-Responders to Oncology Therapeutics

Are Highly Prevalent and Very Costly

BRAF Inhibitor Shrinks Metastatic Melanoma

But ONLY in patients whose tumors have the BRAF mutation.Only a subset of patients have a durable response.

BRAF Inhibitor Prolongs Survival in Patients with Metastatic Melanoma

McDermott U et al. N Engl J Med 2011;364:340-350.

Immunotherapy: No Silver Bullet

But despite the clinical success of antibodies against the immune

regulators CTLA4 and PD-L1/PD-1, only a subset of people exhibit

durable responses, suggesting that a broader view of cancer

immunity is required.

“Even within NCCN, certainly the majority of decision

nodes that are enshrined in NCCN [guidelines] are not

supported by high level evidence.”

Dr. Clifford Hudis

President, ASCO

Interview in Cancer Letter 22 Nov. 2013, 39

Evidence: We Still Need Research

NCCN = National Cooperative Cancer Center Network

The Solution Must Be Driven

by Data

We Need to Invest

in Big Data Research

National Research Council on Big DataNational Research Council’s 2013 “Frontiers of Massive

Data Analysis” report concluded:

• The challenges associated with massive data go

far beyond the technical aspects of data.

• The key element in meeting Big Data’s challenges

is the development of rigorous quantitative and

statistical methods.

• It is quite possible to turn data into something

resembling knowledge when actually it is not.

• Overlooking this foundation may yield results that

are, at best, not useful, or harmful at worst.

http://www.nap.edu/catalog.php?record_id=18374

Key Challenges in Big Data

• Preprocessing (Normalization) and Hot Spot Detection

• Need methods to compare measurements across sources

and to rapidly identify salient features

• Data Integration

• Need methods that can combine data from various

sources where there are hidden correlations in the data

• Reproducible Research

• Need to leverage the volume and velocity of the data to

provide opportunities for validation of findings

• Network Methods

• Need to move beyond correlations in studying

relationships in data

• Data Access and Utility

• Need to assure that different constituencies have access

to data in a form that meets their unique needs

Network Methods

Integrative Network Inference:

PANDA

Kimberly Glass, GC Yuan

Gen

es

Condition

s

Expression data

(Angiogenic)

Gen

es

Condition

s

Expression data

(Non-angiogenic)

Co

mp

are

/Iden

tify D

iffere

nces

Network for

Angiogenic Subtype

Network for

Non-angiogenic Subtype

PANDA: Integrative Network Models

Kimberly Glass, GC Yuan

Kimberly Glass, GC Yuan

Kimberly Glass, GC Yuan

Regulatory Patterns suggest Therapies

good

prognosis

poor

prognosis

• Single-sample networks for TCGA

glioblastoma patients

• 3yr survival to define good and poor

prognosis

• LIMMA analysis using network

“edge Z-scores”

• Analysis points to important roles

for FOS-JUN and NFkB

• Gene degree differences identify

mitosis and immune-related genes

Significant

differential targeting

(gene degree

differences),

GSEA, FDR<0.01

higher targeting in

good

higher targeting in

poor

Glioblastoma

network signatures

With Big Data comes Big Opportunity

• We are at a point in time where data generation is a

commodity.

• Data management, data provenance, data cleaning, data

wrangling are ongoing challenges.

• Solving these problems provides opportunities to drive

new discoveries important in health and biomedical

research.

• In pharma, there are tremendous opportunities for drug

repurposing, drug discovery, and drug rescue if you

develop integrated data strategies and intelligent

analyses.

• For pharma, part of the strategy for success is

recognizing that you are a data company.

Before I came here I was confused

about this subject.

After listening to your lecture,

I am still confused but at a higher level.

- Enrico Fermi, (1901-1954)

<johnq@jimmy.harvard.edu>

Acknowledgments

DFCI GTEx Team

Joseph Barry

Joey (Cho-Yi) Chen

Shelia Gaynor

Marieke Kuijjer

Camila Lopes-Ramos

Megha Padi

Joseph Paulson

John Platig

John Quackenbush

Daniel Schlauch

Administrative

Support

Nicole Trotman

http://compbio.dfci.harvard.edu

CDNM, Brigham and

Women’s Hospital

Dawn DeMeo

Kimberly Glass

Abhijeet Sonawane