Supporting Visual Comparison and Pattern Identification in Widescale Genomic Datasets

Venkat Bandi, Carl Gutwin

Keywords: Interactive visualization, genomic visualization, coordinated and multiple views, genotype map, haplotype map

TL;DR: Design of a novel visualization tool for exploring widescale genomic datasets through new interaction techniques and visual representations.

Abstract: Large-scale linear datasets are often visualized using tables. Visual analysis tasks in such systems involve comparisons and identification of patterns across rows and columns, but these tasks can be hard to perform as the table increases in size because rows and columns of interest can be far apart in the table. This problem is particularly evident in table visualizations of genomic data such as SNPs (which are genetic markers used in comparing different variants of an organism). Visual analysis of SNP datasets has a wide range of applications in plant breeding, genome-wide association studies, and pharmacogenetics. However, current SNP visualizations are limited in their support for complex analytic tasks in wide-scale tables. Through ongoing collaborations with genomic researchers and plant breeders, we have identified a set of new interaction requirements for visual analysis of SNP datasets, and we have developed a new visualization tool with interaction techniques that satisfy the requirements. Our requirements and techniques provide new understanding of how to support complex visual analysis in large-scale table visualizations.

Track: HCI/visualization

Revision: No

Revision Reviewers: No opinion

Summary Of Changes:

Based on the recommendation of the meta-review and feedback from other reviewers, we have made the following changes to our document (organized below based on the main requests in the meta-review):

1. Provide annotations for the figures

- We have updated our main figure (Figure 1) with annotations (a-e) and have expanded the figure caption to reference the annotations. The caption now provides a more detailed description of each part of the system shown in Figure 1.
- We reviewed each of the other figures to ensure that the features referenced by the captions were clearly indicated in the images. In some of the figures, there were sufficient visual landmarks that we did not feel that adding annotation marks would improve clarity. Figure 8 also contains annotations and highlights, and we used a consistent annotation scheme between Figure 1 and Figure 8.

2. Fix the stylistic issues with the writing that are raised by the reviewers

- Thanks to the reviewer for their feedback on this issue! We have addressed the excessive use of "i.e." and "e.g." and corrected them accordingly. We have replaced these abbreviations in most instances with simpler language; we have retained a few usages where necessary (but the total number is dramatically reduced).
- In addition, we have done a full edit pass on the paper to improve flow and fix other minor grammatical issues.

3. Be more specific about the experts who have used (and currently are using) the system

- We have added in additional information regarding our collaborators, and have revised the text to better explain how they use our system for their research work. We have also cited papers that have used our visualization system, and have added references to plant-breeding websites that have incorporated our tool.
- We now provide a link to the live system and the source code repository.
- We have also included a video that shows the system in action, including all the new interaction features.

4. Refer to your design requirements (R1 - R6) in your system overview and method descriptions

- We have revised the system overview section to explicitly refer back to the design requirements that we set out earlier in the paper.

Thanks again for the reviews - these revisions have improved the paper's clarity and real-world relevance.