Interactive Visualization Tools

Organizers: Richard Payne (Lilly), Freda Cooner (Amgen)
Chair: Richard Payne (Lilly)
Vice Chair: Fang Chen (SAS)

Vital Reddy Jaggavarapu (Boehringer-Ingelheim)
Ajay Gupta (PPD)
Stefan Avey (J&J)
Jocelyn Sendecki (J&J)
Jeremy Wildfire (Gilead)


Title: Oncology Data Analytics and Insights Rshiny Application
Speaker: Vital Reddy Jaggavarapu (Boehringer-Ingelheim)

To make the appropriate scientific data-based decisions, visualization tools and technologies are essential to analyze massive amounts of data and stay focused on the relevant information. R shiny will help with advanced data analytics and interactive visualization for clinical data. Oncology clinical trial data oversight using the static reports has always been challenge or achieve clarity , as it does not allow the user to have a flexibility to explore the data in detailed manner to get a better understanding of the data. It also creates user to depend on the technical experts, who support to generate the reports. The whole process has been time and resource intensive and less efficient. With more and more interactive visualization tools availability in the industry, we have decided to use one of the tool to enable us to solve aforementioned problem. R-shiny enabled us to create an app, which can be utilized by a user, just by providing the access to the input data to the app. This app can process data and produce interactive reports and plots that can help real time monitoring of data during the conduct of the clinical trial. This talk will present high-level overview of the above-mentioned R-shiny app.

Title: Advanced Visualization using TIBCO Spotfire® and SAS® using SDTM Data
Speaker: Ajay Gupta (PPD)

Stakeholder requests for quick access to clinical research data to explore data interactively are increasingly common in the pharmaceutical industry. TIBCO Spotfire is an analytics and business intelligence platform which enables data visualization in an interactive mode. Users can further integrate TIBCO® Spotfire with SAS® and create visualizations with powerful functionality that includes data filters and data flags. These visualizations can help the user to self-review the data in multiple ways resulting in significant time and cost advantages. This presentation will demonstrate some advanced visualizations from Preclarus® Patient Data Dashboard (Preclarus PDD) within PPD® created using TIBCO Spotfire and SAS (for the SDTM database) and share our experiences and challenges while creating this visualization.

Title: Visualizing Activity Data from Wearable Devices in R to Create Personalized Activity Reports
Speaker: Stefan Avey (J&J)

Wearable devices, such as wrist-worn activity trackers, are increasingly being used in observational and interventional clinical studies to assess physical functioning and sleep quality in an individual’s natural environment. These devices contain accelerometers which typically sample data 60-100 times per second. These raw data are typically further summarized into epoch-level (e.g., 1 min) activity counts which are further summarized into daily endpoints (e.g. total sleep time). In certain studies, it may be desirable to share summaries of activity data back with participants to increase engagement and compliance. This talk will present a case study on the use of R Shiny to allow interactive creation of personalized activity reports using the R ggplot2 package for visualization. The design elements that informed the content and style of visualizations will be the primary focus and important technical implementation details will be highlighted.

Title: Molecular Candidate Characterization in High-Throughput Biology Using Interactive Graphics in R
Speaker: Jocelyn Sendecki (J&J)

The generation of hundreds or thousands of candidate molecules is an integral part of early drug development. Molecular characterization at this stage involves a variety of high-throughput biological assays; this presentation will focus on specifically epitope binning with regard to monoclonal antibodies.
Epitope binning characterizes antibodies by their antigen binding site, a critical piece of information as the binding site plays a pivotal role in downstream cellular response. The output data generated by this assay is a binary matrix which reflects all pairwise relationships between antibodies in a batch. This matrix is then used to inform bin membership through heatmaps, hierarchical clustering, and community network analysis. We have developed an R Shiny app to support this determination using multiple coordinated approaches.
In this context, three aspects of data visualization will be highlighted: converting static graphics to interactive graphics to facilitate visualization data otherwise too large to display, interactive cluster selection and visualization, and three-dimensional graphics.

Title: Building Open Source Tools for Safety Monitoring: Advancing Research Through Community Collaboration
Speaker: Jeremy Wildfire (Gilead)

The Interactive Safety Graphics (ISG) workstream of the ASA-DIA Biopharm Safety Working Group is excited to introduce the safetyGraphics package: an interactive framework for evaluating clinical trial safety in R using a flexible data pipeline. Our group seeks to modernize clinical trial safety monitoring by building tools for data exploration and reporting in a highly collaborative open source environment.  At present, our team includes clinical and technical representatives from the pharmaceutical industry, academia, and the FDA, and additional contributors are always welcome.  The current release of the safetyGraphics R package includes graphics related to drug-induced liver injury. The R package is paired with an in-depth clinical workflow for monitoring liver function created by expert clinicians based on medical literature. safetyGraphics features interactive visualizations built using htmlwidgets, a Shiny application,  and the ability to export a fully reproducible instance of the chart with associated source code. To ensure quality and accuracy, the package includes more than 300 unit tests, and it has been vetted through a beta testing process that included feedback from more than 20 clinicians and analysts.  The Shiny application can easily be extended to include new charts or applied to other disease areas due to its modular design and generalized charting framework. Several companies have adapted the tool for their own use, leading to interesting discussions and paving the way for enhancements, which demonstrates the power of open source and community collaboration.