As our ability to generate big data continues to increase, from DNA sequencing to electronic health record and imaging data, multifaceted data extraction and analysis has become the next challenge in genomics research, drug discovery, and clinical development. Track 8 aims to address a variety of the latest visualization techniques and tools, including how to design, implement, and evaluate them, for each individual field.

Tuesday, April 5

7:00 am Workshop Registration and Morning Coffee

2:00 – 6:00 Main Conference Registration

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

Wednesday, April 6

7:00 am Registration Open and Morning Coffee

9:00 Benjamin Franklin Awards and Laureate Presentation

9:30 Best Practices Awards Program

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

DATA VISUALIZATION FOR GENOMIC DATA AND DRUG DISCOVERY

10:50 Chairperson’s Opening Remarks

Tom Johnstone, Managing Partner, Health & Life Sciences, Knowledgent

11:00 Approaches for the Integration of Visual and Computational Analysis of Biomedical Data

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

The integration of computational and statistical approaches with visualization tools is becoming crucial as biomedical data sets are rapidly growing in size. Finding efficient solutions that address the interplay between data management, algorithmic and visual analysis tools is challenging. I will discuss some of these challenges and demonstrate how we are addressing them in our Refinery Platform project (http://www.refinery-platform.org).

11:30 FireBrowse: Mining the Firehose of TCGA Genomic Data

Michael Noble, Assistant Director for Data Science, Cancer Genome Analysis, Broad Institute

We introduce FireBrowse, a companion portal to the Broad Institute GDAC Firehose analysis pipeline. Developed for The Cancer Genome Atlas, and backed by a powerful compute infrastructure, programming interface, online reports and modern graphical tools, FireBrowse provides a simple yet capable means of visually and programmatically exploring one of the most comprehensive and deeply characterized open cancer datasets in the world.

12:00 pm Bringing Modeling to the Masses: LiveDesign, a Platform for Collaborative Drug Discovery

Sol Reisberg, Director, Business Development for Enterprise Informatics, Schrödinger

While the scientific power of computational chemistry has dramatically increased over the last several years, usage of computational methods in industry remains limited because of difficulties associated with software usage. Here, we present LiveDesign, a tool designed to allow collaborative design of small-molecule compounds, and rapidly prioritize designed ideas based on computational feedback. The thin client provides users with extreme ease-of-use, while the cloud-hosted server is highly scalable.

12:30 Session Break  

12:40 Luncheon Presentation I (Sponsorship Opportunity Available)

1:10 Luncheon Presentation II (Sponsorship Opportunity Available)

1:40 Session Break

1:50 Chairperson’s Remarks

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

1:55 IOBIO: Interactive, Visually-Driven, Real-Time Analysis of Genomic Big Data

Chase Miller, Director of Research and Science, University of Utah, Eccles Department of Genetics, USTAR Center for Genetic Discovery, University of Utah School of Medicine

IOBIO is a web-based system for big genomic data, which uses visualization coupled with real-time analysis to better understand complex and opaque data. We have developed several IOBIO web apps including quality control analysis of genomic alignment and variant data, interrogation of potential disease causing variants, and species identification and classification of raw sequencing data.

2:25 Big Mechanism Visualization: Interactive Analysis Techniques for Understanding Biological Pathway Networks

Angus Forbes, Assistant Professor, Computer Science, University of Illinois at Chicago

Understanding causality in biological pathways remains an active area of research for systems biologists, cancer researchers, and drug designers. This talk discusses recent explorations of interactive techniques that enable visual analysis tasks related to representing and analyzing causality in pathway networks, including identifying feedback loops and simulating the downstream effects of perturbing networks by “knocking out” proteins or protein complexes.

2:55 Innovation through Information: Enabling Proactive Healthcare Outcomes

Chris Blotto, Managing Partner, Knowledgent

Today, methods to collect information about health and disease state are more advanced than ever. Modern day devices such as wearables and tablets, allow us to capture real world data that’s now being combined with data sets such as cross-study clinical outcomes data, genetic biomarkers and compound/biologic target data to perform advanced analytics that are truly game changing. These methods are reducing the time it takes to execute analytic research projects from months, to days or hours. This discussion will focus on introducing the audience to architectures, technologies and models that have proven successful in this capacity.

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

TOOLS FOR EXPLORING AND VISUALIZING IMAGING DATA

4:00 Automating Image-Based High Content Screening

Fethallah Benmansour, Ph.D., Senior Imaging Specialist, Pharma Research & Early Development Informatics (pREDi), Roche Innovation Center Basel

Our integrated solution allows for automated data processing, on-the-fly interactive data mining and data visualization. By linking the data points to the images in a dynamically adjustable fashion, the solution allows for efficient QCing of the high content screening processes (including image analysis). It simplifies the study summary reports providing more confidence on the scientific findings.

4:30 BugID: An Intelligent Recognition System for Storage Pest Fragments Contaminating Food Products

Joshua Z. Xu, Ph.D., Senior Computer Scientist, Division of Bioinformatics and Biostatistics, National Center for Toxicological Research, U.S. Food and Drug Administration

Species identification of food contaminating insect fragments is critical to FDA’s risk analysis and decision making during safety inspection of FDA-regulated food products. Combining image analysis and machine intelligence techniques, BugID will increase the reliability and throughput of food inspection by providing fast, consistent, and accurate insect identification results.

5:00 Enable Cancer Immunotherapy via Integrative Tissue Analytics

Franziska Mech, Ph.D., Data Scientist, Pharma Research and Early Development, pRED Informatics, Roche Innovation Center Penzberg

Establishing automated tissue imaging as high-throughput tool for understanding tissue context in the era of cancer immunotherapy. Integrating the obtained imaging data with other data sources such as clinical and genomic information and making it available for data scientists and biomarker experts via tailored interactive visualization tools.

5:30 – 6:30 Best of Show Awards Reception in the Exhibit Hall with Poster Viewing

Thursday, April 7

7:00 am Registration and Morning Coffee

10:00 Coffee Break in the Exhibit Hall and Poster Competition Winners Announced

VISUALIZING ELECTRONIC HEALTH RECORDS FOR CLINICIANS

10:30 Chairperson’s Opening Remarks

Tom Johnstone, Managing Partner, Health & Life Sciences, Knowledgent

10:40 A Real-Time Data-Driven Visualization within the Electronic Health Record

Randi Foraker, Ph.D., M.A., Assistant Professor, Epidemiology, College of Public Health, The Ohio State University

Health visualizations at the point-of-care can help bring electronic health record data to life for the patient and the provider. Automated tools that provide such visualizations can enhance patient-provider communication and shared decision-making, and make the healthcare encounter more efficient.

11:00 Information Visualization for Cognitively Guided Chronic Disease Risk Assessment and Personalized Interventions

Rema Padman, Professor of Management, Science & Healthcare Informatics, The H. John Heinz III College, Carnegie Mellon University

This presentation describes a novel methodology and a prototype software tool for quantitatively summarizing and visually displaying contextualized information on many relevant risk factors across many patients, which is particularly appropriate for chronic disease risk assessment. Using statistical dimensionality reduction methods combined with a novel data visualization approach, the tool provides two-dimensional visualizations and binary classification of chronic disease risk.

11:20 Visualizing Big Data and The Future of Cancer Care

Andrew K. Stewart, MA, Chief, Oncology Data, CancerLinQ

CancerLinQ is an clinical quality of care initiative of the American Society of Clinical Oncology. Data visualization for practicing oncologists is a cornerstone of the project. This presentation will illustrate approaches to visualizing quality of care performance metrics, longitudinal patient time-lines, and facilitating ad-hoc data interrogation by clinical users.

11:40 Raising the Bar for Central Medical Review

Victor Lobanov, Ph.D., Executive Director, Data Sciences, Covance Inc

Periodic review of clinical data is critical for the patient safety and data quality. Covance’s Medical Review is aligned with the FDA guidance for a greater role of central monitoring and provides timely, integrated views of all relevant clinical data along with the unique, interactive capabilities to detect outliers and trends, create and analyze cohorts, execute review workflows, annotate clinical data, and communicate observations.

12:10 pm Session Break

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

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

DATA VISUALIZATION FOR CLINICAL AND TRANSLATIONAL DATA

1:55 Chairperson’s Remarks

Nirmal Keshava, Ph.D., Senior Principal Informatics Scientist, Research & Development Information, AstraZeneca PLC

2:00 Delivering Standardized Clinical and Preclinical Data to Investigators in Guided Visualization Using Spotfire 6.5

Baisong Huang, Principal Statistical Analyst, NIBR Informatics, Novartis Institutes for Biomedical Research

As visualization tools evolve and become widely accepted in investigating and monitoring drug safety and efficacy, rapid access to standardized, interpretable data views is becoming essential. We will present some examples how we standardized and aggregated data in both translational and clinical settings and provided guided analysis to visualize the data in real time.

2:30 Deriving Knowledge from Real-World Evidence Using Large-Scale Analytics

Nirmal Keshava, Ph.D., Senior Principal Informatics Scientist, Research & Development Information, AstraZeneca PLC

In this talk, I will discuss the effort to develop large-scale analytics to derive knowledge and value from real-world evidence. This will be done in the context of using clinical data in real-world evidence databases to answer critical questions that can arise in both the clinical and pre-clinical problem spaces. I will focus on defining how the business problem is accurately translated into a mathematical problem and how that problem is addressed by data from real-world evidence databases.

3:00 Instrumenting the Healthcare Enterprise for Discovery Research

Shawn Murphy, M.D., Ph.D., Director, Research Computing and Informatics, Partners Healthcare; Associate Professor, Neurology, Harvard Medical School; Associate Director, Laboratory of Computer Science, Massachusetts General Hospital

The Healthcare Enterprise produces enormous amounts of data during clinical care that could potentially be used for human research. However, the quality of the data is very raw, and privacy concerns are paramount. Deriving knowledge from the data requires a combination of searching the data visually for hypotheses, computing derived patient attributes with well understood accuracies, and obfuscating data when necessary to preserve patient privacy.

3:30 Visualizing Variability in Electronic Health Records: The Variability Explorer Tool (VET)

Hossein Estiri, Ph.D., Senior Fellow, Institute of Translational Health Sciences, University of Washington

This presentation describes application of visual analytics in development of the Variability Explorer Tool (VET), which is designed to detect and explore variability in Electronic Health Records (EHR) data. Existing variability in EHR data limits their utility for healthcare decision-making and research. VET provides a suit of open-source statistical solutions to detect and explore variability across time and between units of analysis in EHR data.

4:00 Conference Adjourns