Context-Driven Narrative Visualizations for "Big Data" Applications in Medical Biology

Medical biology data have grown exponentially and are more complex than our minds can process. It is common for omics datasets obtained from measuring molecular entities (e.g., genes in genomics and proteins in proteomics) to undergo open-ended analysis due to the complexity of biological systems. Rather than probing for a single answer, researchers often explore the data to discover multiple new insights, form actionable next steps, and frame new questions. However, existing visualizations are limited in their ability to 1) help understand the details of the molecular entities that they contain; 2) provide clues of how to move forward; and 3) show which structural biological entities are relevant. We describe how context-driven narrative visualizations with guidance can address these limitations. We describe a prototype, visAPPprot, as an example.