Cancer Informatics

The Cancer Informatics track explores the important technology, informatics trends, and challenges of applying computational biology to cancer research and care. Themes that will be covered in expert-led presentations include: collaboration and network models; data access/management/integration strategies; and applications of biological interpretation to aid in research at the benchside or care at the bedside. Most clinical diagnoses involve the use of clinical testing, much of which is not standardized locally/nationally/internationally. How do your approaches address this reality?

Final Agenda

Monday, April 20

9:00 am - 5:00 pm Hackathon*

*Pre-registration required.

Tuesday, April 21

7:30 am Workshop Registration Open and Morning Coffee

8:30 am - 3:30 pm Hackathon*

*Pre-registration required.


8:30 - 11:30 am Recommended Morning Pre-Conference Workshops*

W3. Introduction to Data Visualization for Biomedical Applications

Nils Gehlenborg, PhD, Assistant Professor, Department of Biomedical Informatics, Harvard Medical School

Alexander Lex, PhD, Assistant Professor, SCI Institute, School of Computing, University of Utah

12:30 - 3:30 pm Recommended Afternoon Pre-Conference Workshops*

W12. Cancer Genome Analysis

Jeffrey Rosenfeld, PhD, Manager, Biomedical Informatics Shared Resource and Assistant Professor of Pathology and Laboratory Medicine, Rutgers Cancer Institute of New Jersey; President, Rosenfeld Consulting LLC

*Separate registration required.

2:00 - 6:30 Main Conference Registration Open

PLENARY KEYNOTE SESSION

4:00 Welcome Remarks

Cindy Crowninshield, RDN, LDN, Executive Event Director, Cambridge Healthtech Institute

 

 

 

4:05 Keynote Introduction

4:15 PLENARY KEYNOTE PRESENTATION: NIH’s Strategic Vision for Data Science

Susan K. Gregurick, PhD, Associate Director, Data Science (ADDS) and Director, Office of Data Science Strategy (ODSS), National Institutes of Health

 

 

 

 

Rebecca Baker, PhD, Director, HEAL (Helping to End Addiction Long-term) Initiative, Office of the Director, National Institutes of Health

 

 

 

 

Riffyn_new 5:00 - 7:00 Welcome Reception in the Exhibit Hall with Poster Viewing

 

 

Wednesday, April 22

7:30 am Registration Open and Morning Coffee

PLENARY KEYNOTE SESSION

8:00 Welcome Remarks

Allison Proffitt, Editorial Director, Bio-IT World

 

 

 

8:05 Keynote Introduction

8:15 Toward Preventive Genomics: Lessons from MedSeq and BabySeq

Robert Green, MD, MPH, Professor of Medicine (Genetics) and Director, G2P Research Program/Preventive Genomics Clinic, Brigham & Women’s Hospital, Broad Institute, and Harvard Medical School

 

 

 

8:45 PANEL DISCUSSION: Game On: How AI, Citizen Science, and Human Computation Are Facilitating the Next Leap Forward

Seth CooperSeth Cooper, PhD, Assistant Professor, Khoury College of Computer Sciences, Northeastern University

 

 

 

 

 

Lancashire_LeeLee Lancashire, PhD, Chief Information Officer, Cohen Veterans Bioscience

 

 

 

 

 

Pietro Michelucci, PhD, Director, Human Computation Institute

 

 

 

 

 

Jérôme WaldispühlJérôme Waldispühl, PhD, Associate Professor, School of Computer Science, McGill University

 

 

 

 

 

While the precision medicine movement augurs for better outcomes through targeted prevention and intervention, those ambitions entail a bold new set of data challenges. Various panomic and traditional data streams must be integrated if we are to develop a comprehensive basis for individualized care. However, deriving actionable information requires complex predictive models that depend on the acquisition and integration of patient data on a massive scale. This picture is further complicated by new data streams emerging from quantified self-tracking and health social networks, both of which are driven by experimentation-feedback loops. Tackling these issues may seem insurmountable, but recent advancements in human/AI partnerships and crowdsourcing science adds a new set of capabilities to our analytic toolkit. This talk describes recent work in online collective systems that combine human and machine-based information processing to solve biomedical data problems that have been otherwise intractable, and an information processing ecosystem emerging from this work that could transform the landscape of precision medicine for all stakeholders.

9:45 Coffee Break in the Exhibit Hall with Poster Viewing

PRECISION CANCER MEDICINE METHODS

10:50 Organizer’s Welcome Remarks

Cambridge Healthtech Institute

10:55 Chairperson’s Remarks


11:00 KEYNOTE PRESENTATION: Using Networks to Understand Genetic and Genomic Drivers of Disease

Quackenbush_JohnJohn Quackenbush, PhD, Henry Pickering Walcott Professor of Computational Biology and Bioinformatics; Chair, Department of Biostatistics, Harvard T.H. Chan School of Public Health

This presentation will address the problem of biological complexity in which many factors, each of small effect size, collectively influence disease risk, development, complexity, and response to therapy in cancer and other complex diseases. By using innovative computational methods built around network representations of biological interactions, we can gain insight into the disease process, develop predictive biomarkers, and identify possible avenues of therapeutic intervention.

11:30 Precision Cancer Medicine

Rosenfeld_JeffreyJeffrey Rosenfeld, PhD, Manager, Biomedical Informatics Shared Resource and Assistant Professor of Pathology and Laboratory Medicine, Rutgers Cancer Institute of New Jersey; President, Rosenfeld Consulting LLC

This presentation will illustrate the current methods that are used for determining the precise treatment of cancer rather than the standard chemotherapy methods.

12:00 pm Sponsored Presentation (Opportunity Available)

PrecisionForMedicine_NoTagline 12:15 pm Presentation to be Announced

12:30 Session Break

12:40 LUNCHEON PRESENTATION I: Advancing Precision Medicine with a Complete Bioinformatics Ecosystem

Davis-Dusenbery_BrandiBrandi Davis-Dusenbery, PhD, CSO, Seven Bridges


1:10 LUNCHEON PRESENTATION II: A Network Polypharmacology Approach to Diffuse Intrinsic Pontine Glioma

MacLean_FinlayFinlay MacLean, MSc, Data Scientist, Elsevier

Network medicine promises to be a potential linchpin in oncological drug repurposing. We developed a multi-scale heterogeneous knowledge graph spanning genomics, epigenetics, transcriptomics and proteonomics. We implemented a random walk and generated dense vector representations of the neighbourhoods (or interactomes) of key nodes and used these in downstream supervised machine learning tasks. Leveraging Entellect we plan to use the models in our collaboration with the University of Zurich, to suggest potential DIPG drug repurposing candidates.

1:40 Session Break

UNDERSTANDING HETEROGENEITY AND HUMAN DISEASE

1:50 Chairperson’s Remarks

Yuval Itan, PhD, Assistant Professor, Department of Genetics and Genomic Sciences; Member, Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai

1:55 Deciphering the Complex Heterogeneity of Cancer

Milos_PatricePatrice Milos, PhD, Co-Founder/President and CEO, Medley Genomics, Inc.

In 2017, 1.7 million people in the US were diagnosed with cancer, and even though cancer survival rates have increased, it still accounts for 1 in 4 deaths annually. Cancer, a heterogeneous disease, has significant tumor cell variability within individual patients, as well as across categories of patients, creating complex barriers to effective and lasting cures for patients. Understanding this heterogeneity will be required to individualize care for patients. Medley Genomics provides a software platform that uses patent-pending algorithms and advanced data analytics to describe a patient’s diverse tumor cell mixture. This enables creation of unique molecular diagnostic fingerprints for improving patient diagnosis, monitoring and treatment of cancer, and helps to improve novel oncology therapies and therapeutic combinations including individual cancer vaccine development.

2:25 Estimating Genotypic Heterogeneity Underlying Human Disease

Itan_YuvalYuval Itan, PhD, Assistant Professor, Department of Genetics and Genomic Sciences; Member, Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai

Whole exome and whole genome sequencing provide hundreds of thousands of genetic variants per patient, however, of them only very few are pathogenic. Current computational methods are inefficient in differentiating pathogenic mutations from neutral genetic variants that are predicted to be damaging and cannot predict the functional outcome of mutations. We will present deep learning approaches and machine learning methods in the role of detecting pathogenic mutations. Visualization tools for better utilizing NGS data will be presented to understand human disease genomics.

2:55 Presentation to be Announced

3:10 Sponsored Presentation (Opportunity Available)

3:25 Refreshment Break in the Exhibit Hall with Poster Viewing

PRIORITIZATION OF DELETERIOUS VARIANTS IN RARE DISEASE

4:00 Chairperson’s Remarks

4:05 Application of Novel Methods and Algorithms for Prioritization of Deleterious Variants in Rare Disease

Worthey_LizElizabeth Worthey, PhD, Director, Genomic Medicine, University of Alabama, Birmingham School of Medicine

When applying genomic methods to identify causal variants for patients with undiagnosed rare disease, the primary goal is to identify one or more deleterious variants resulting in the patient’s clinical presentation. Typically, this is done through filtering and then prioritization of variants with follow up expert review for reporting. Prioritization remains a challenging task due to the complexity of biological networks, the high degree of both phenotypic and molecular variability, and the volume of reference data to be considered. Application of methods that can prioritize variants in the presence of such complexity are clearly of critical importance when the goal is to minimize the manual steps and thus time taken for interpretation. We have developed methods that classify 97% of variants previously human interpreted as pathogenic into the “should be clinically reported” category and are now refining and integrating these into the tools we use for molecular diagnostics at UAB.

GENOME-WIDE DNA METHYLATION PROFILING

4:35 M2A: DNA Methylome Reveals Genome-wide Promoter Activities in Individual Tumors and Captures Tumor-Type-Specific Promoter Usages

Chen_XiangXiang Chen, PhD, Assistant Member, Computational Biology, St. Jude Children’s Research Hospital

Tumors use alternative promoters to increase isoform diversity, to activate oncogenes, and to evade host immune attacks. ChIP-seq has been the gold standard to measure promoter activities, but scarcity of fresh starting materials restricts interrogation of individual patient tumors. We present MethylToActivity (M2A), a deep-learning framework for inference of H3K4me3 and H3K27ac levels from DNA methylomes. M2A accurately reveals promoter activities in individual tumors, captures tumor-type specific promoter usages, and is generalizable to various pediatric and adult cancers.

5:05 Sponsored Presentation (Opportunity Available)

 

Stellus_Technologies

 

 

5:35 Best of Show Awards Reception in the Exhibit Hall with Poster Viewing

 

RedRiver

 

 

 


6:45 End of Day

Thursday, April 23

7:30 am Registration Open and Morning Coffee

PLENARY KEYNOTE SESSION & AWARDS PROGRAM

8:00 Organizer’s Remarks

Cindy Crowninshield, RDN, LDN, Executive Event Director, Cambridge Healthtech Institute

 

 

 

 

8:05 Awards Program Introduction

8:10 Benjamin Franklin Award and Laureate Presentation

J.W. Bizzaro, Managing Director, Bioinformatics.org

 

 

 

 

Discngine8:35 Bio-IT World Innovative Practices Awards

Allison Proffitt, Editorial Director, Bio-IT World

 

 

 

 

9:00 AI in Pharma: Where We Are Today and How We Will Succeed in the Future

Natalija Jovanovic, PhD, Chief Digital Officer, Sanofi Pasteur

 

 

 

 

Penguin_Computing_Tagline 9:45 Coffee Break in the Exhibit Hall and Poster Competition Winners Announced at 10:00

 

 

GENOMIC DATA PROCESSING AND APPLICATION: TOOLS AND PIPELINES

10:30 Organizer’s Remarks

Cambridge Healthtech Institute

10:35 Chairperson’s Remarks

10:40 A Deep Dive on DRAGEN-GATK

Van_der_Auwera_GeraldineGeraldine Van der Auwera, PhD, Director of Outreach and Communications, Broad Institute of MIT and Harvard

This presentation will provide visibility into the future of the GATK Best Practices, the most widely used genomic data analysis pipeline, and forthcoming DRAGEN-GATK release, which will be made available in open source form as well as a licensed hardware accelerated version. A deeper understanding of the contents, performance, and accuracy of the DRAGEN-GATK Best Practice pipeline, updates on when they can expect the tools to be released, and how they can leverage DRAGEN-GATK in their own research once it becomes available.

11:10 Making RWD Actionable for R&D and Cancer Care

Singal_GauravGaurav Singal, MD, Chief Data Officer, Foundation Medicine, Inc.

As cancer sequencing becomes increasingly common, growing real-world datasets are emerging which hold the promise of accelerating the pace of discovery and improving patient care. However, numerous actionability gaps remain in deploying RWD and cancer genomics to these domains. In this talk, we’ll explore, in the context of datasets that have been developed and deployed, how problems facing drug developers and oncologists may be tackled now and in the future.

11:40 Genomic Results as Discrete Searchable Data in the Electronic Health Record

Skelton_TimothyTim Skelton MD PhD, Medical Director, Core Laboratory & Laboratory Informatics, Lahey Hospital & Medical Center

Dr. Skelton will describe how Lahey Hospital & Medical Center’s precision medicine program has achieved a closed loop integrated system for discrete data capture and storage into Epic EHR. Participants will learn how collaboration among data scientists, patient care providers, and computer analysts is being used to change cancer care using decision support tools and business intelligence. The dependency on the discrete data capture of all relevant genomics information, including order, procedure, result, and clinical, into a single relational database will be emphasized. Skills required to process and deliver information in a manner that effectively changes patient care in the direction of high value will be described.

12:10 pm Session Break

12:20 Luncheon Presentation (Sponsorship Opportunity Available) or Enjoy Lunch on Your Own

1:20 Dessert Refreshment Break in the Exhibit Hall with Last Chance Poster Viewing

APPLICATION OF TOOLS IN CLINICAL, POPULATION, AND COLLABORATIVE HEALTH SETTINGS

1:55 Chairperson’s Remarks

Chris Anderson, Editor-in-Chief, Clinical OMICs

2:00 Open-AD: A Radically Open Approach to Diversify the AD Drug Portfolio

Mangravite_LaraLara Mangravite, PhD, President, Sage BioNetworks

Here we describe a radically open approach to diversify the AD drug portfolio. Using multi-omic and genetic models of disease built from human brain data, a suite of emerging therapeutic hypotheses are generated that complement the small set already in drug development. To catalyze rapid evaluation of these targets, target enabling packages – containing computational and experimental resources including prototype drug compounds – are developed and openly distributed for use across the research community.

2:30 Development of Risk Prediction Models for Cardiovascular Diseases and Prostate Cancer Using Deep Learning: Case Studies from Ongoing Collaboration between the Department of Veterans Affairs (VA) and the Department of Energy (DOE)

Madduri_RaviRavi Madduri, Scientist, Data Science and Learning, Argonne National Laboratory; Senior Scientist, University of Chicago Consortium for Advanced Science and Engineering (UChicago CASE)


3:00 A Deluge of Data, Wrangling the UK BioBank and Other NGS Data

Sasson_AriellaAriella Sasson, PhD, Senior Research Investigator, Bristol-Myers Squibb

As the amount of NGS data grows, the basic standard manipulations of data (movement, storage, organization) become non-trivial problems (cost, time, space, corruption). These data, which can grow to petabytes in size, not only bring the question of how we handle the data to the forefront but compound the difficult task of making it maximally useful and accessible to enable the science.

3:30 Open Science with OHDSI: From Question to Evidence in 5 Days

Bochove_Kees_vanKees van Bochove, Founder, The Hyve


4:00 Close of Conference



Platinum Sponsors

accenture

BenchlingNEW

Elsevier-square

L7-informatics

linguamatics

Nutanix

PerkinElmer

Weka