A Unified Approach to Exploration, Authoring, and Communication with Reproducible Visualizations
1 like246 views
The document discusses a unified approach to exploration, authoring, and communication through reproducible visualizations in data analysis and biomedical informatics. It highlights the significance of data visualization in addressing reproducibility issues in science and illustrates its applications in discovering tumor subtypes and other clinical insights. The concept of 'vistory' is introduced, blending visualization, storytelling, and history to enhance collaborative data analysis and communication.
A Unified Approach to Exploration, Authoring, and Communication with Reproducible Visualizations
- 1. Nils Gehlenborg, PhD Department of Biomedical Informatics Harvard Medical School http://gehlenborglab.org @ngehlenborghttp://gehlenborglab.org A Unified Approach to Exploration, Authoring, and Communication with Reproducible Visualizations
- 5. http://gehlenborglab.org Machine Human Data Methods for Data Visualization and Exploration Tools for Reproducible Research
- 6. What are applications of data visualization?
- 12. Image Editing ToolCytoscape Insight EXPLORATION EXPLANATION
- 13. Image Editing ToolCytoscape DATA PIXELS
- 14. Image Editing ToolCytoscape DATA PIXELS
- 15. DATA PIXELS Dead wood? Hard to move Keeps on giving! Best part Easy to share Low-hanging fruit?
- 16. Why should we care?
- 17. Nature asked 1,576 researchers if there is a reproducibility crisis in science. M Baker, Nature 533, 452-454, 2016
- 18. 0% 100% No crisis (3%) Don’t know (7%) Slight crisis (38%) M Baker, Nature 533, 452-454, 2016 Significant crisis (52%) Nature asked 1,576 researchers if there is a reproducibility crisis in science.
- 19. M Baker, Nature 533, 452-454, 2016
- 20. Intentional? Inability to capture everything? Inability to communicate everything? M Baker, Nature 533, 452-454, 2016
- 21. Intentional? Inability to capture everything? Inability to communicate everything? SOCIAL ISSUE TECHNICAL ISSUES M Baker, Nature 533, 452-454, 2016
- 22. Why does this matter for visualization?
- 23. Discovery of Tumor Subtypes PROBLEM 1 Visualize overlap of patient sets across two or more stratifications. PROBLEM 2 Visualize characteristics of patient sets within a stratification of interest. StratomeX: Exploratory Data Visualization M Streit, A Lex, S Gratzl, C Partl, D Schmalstieg, H Pfister, P Park, N Gehlenborg, Nature Methods (2014)
- 24. M Streit, A Lex, S Gratzl, C Partl, D Schmalstieg, H Pfister, P Park, N Gehlenborg, Nature Methods (2014) Discovery of Tumor Subtypes
- 26. PROBLEM 3 Identify relevant stratifications, pathways, and clinical variables. Discovery of Tumor Subtypes PROBLEM 1 Visualize overlap of patient sets across two or more stratifications. PROBLEM 2 Visualize characteristics of patient sets within a stratification of interest. StratomeX: Exploratory Data Visualization M Streit, A Lex, S Gratzl, C Partl, D Schmalstieg, H Pfister, P Park, N Gehlenborg, Nature Methods (2014)
- 27. Is there a mutation that overlaps with this mRNA cluster? Is there a CNV that affects survival? Is there a pathway that is enriched in this cluster? Query Stratifications Clinical Params Pathways Guided Exploration M Streit, A Lex, S Gratzl, C Partl, D Schmalstieg, H Pfister, P Park, N Gehlenborg, Nature Methods (2014)
- 28. Query Rank Visualize Stratifications Clinical Params Pathways Guided Exploration M Streit, A Lex, S Gratzl, C Partl, D Schmalstieg, H Pfister, P Park, N Gehlenborg, Nature Methods (2014)
- 30. ?
- 33. And now what?
- 37. finding figure/videoAuthoringExploration Presentation DATA-DRIVEN COMMUNICATION finding figure/videoAuthoringExploration Presentation Current Model DATA-DRIVEN DISCOVERY
- 38. DATA-DRIVEN COMMUNICATION finding figure/videoAuthoringloration Pre What we show. DATA-DRIVEN DISCOVERY
- 39. DATA-DRIVEN COMMUNICATION finding figure/videoAuthoringloration Pre What we show. What we tell. DATA-DRIVEN DISCOVERY
- 40. DATA-DRIVEN COMMUNICATION finding figure/videoAuthoringloration Pre What we show. What we did. DATA-DRIVEN DISCOVERY
- 41. DATA-DRIVEN COMMUNICATION finding figure/videoAuthoringloration Pre What we show. What we did. DATA-DRIVEN DISCOVERY
- 42. CLUE vistories Authoring Exploration Presentation Vistories DATA-DRIVEN COMMUNICATION DATA-DRIVEN DISCOVERY
- 51. VISTORY = visualization + story + history
- 52. VISTORY = visualization + story + history Do collaborative data analysis. Use during peer-review. Publish with a paper. Embed in a presentation.
- 53. 49
- 54. 50
- 55. H Stitz et al., to appear in Proceedings of VAST 2018 Marc Streit Lab: Vistories in Action Retrieval and Analysis of Visualization Provenance
- 56. H Stitz et al., work in progress Marc Streit Lab: Vistories in Action Vistories integration with Jupyter Notebooks
- 61. DATA-DRIVEN DISCOVERY DATA-DRIVEN COMMUNICATION http://vistories.org SAMUEL GRATZL ALEXANDER LEX MARC STREIT