Tasks, Techniques, and Tools for Genomic Data Visualization (1905.02853v1)

Abstract: Genomic data visualization is essential for interpretation and hypothesis generation as well as a valuable aid in communicating discoveries. Visual tools bridge the gap between algorithmic approaches and the cognitive skills of investigators. Addressing this need has become crucial in genomics, as biomedical research is increasingly data-driven and many studies lack well-defined hypotheses. A key challenge in data-driven research is to discover unexpected patterns and to formulate hypotheses in an unbiased manner in vast amounts of genomic and other associated data. Over the past two decades, this has driven the development of numerous data visualization techniques and tools for visualizing genomic data. Based on a comprehensive literature survey, we propose taxonomies for data, visualization, and tasks involved in genomic data visualization. Furthermore, we provide a comprehensive review of published genomic visualization tools in the context of the proposed taxonomies.

• The paper establishes taxonomies for genomic data visualization, linking algorithmic analysis with cognitive interpretation.
• It reviews 83 tools categorized by layout and arrangement, enhancing the analysis of complex genomic datasets.
• The study outlines future opportunities in managing 3D genome interactions and integrating genomics with biomedical data.

Overview of Genomic Data Visualization: Tasks, Techniques, and Tools

The paper, authored by Nusrat, Harbig, and Gehlenborg, provides an in-depth examination of genomic data visualization, focusing on the critical tasks, techniques, and tools that have emerged over the past two decades. Genomic data visualization is indispensable for the interpretation and hypothesis generation in genomics, bridging the gap between algorithmic approaches and cognitive skills required for data comprehension.

Genomic Data Visualization Challenges and Developments

The advent of affordable high-throughput sequencing technologies has made the generation and handling of genomic data a cornerstone in biological and medical research. However, the sheer volume and complexity of this data present significant challenges, including the discovery of unexpected patterns and hypothesis formulation from vast genomic landscapes. This necessity has spurred the development of numerous data visualization techniques and tools. The paper postulates taxonomies for the data types, visualization methodologies, and tasks associated with genomic data visualization to provide a structured framework for the field.

Taxonomies and Visualization of Genomic Data

The paper introduces three primary taxonomic structures:

1. Data Taxonomy: Characterizing genomic features, the taxonomy distinguishes between point and segment features, alongside sparse and contiguous feature sets. The taxonomy also emphasizes feature attributes, interconnectivity (within and between genomic sequences), and associated metadata.
2. Visualization Taxonomy: This taxonomy categorizes visualization techniques based on layout, abstraction, partition, and arrangement of sequence axes. Visualization methodologies are organized into linear, circular, space-filling, and spatial layouts, with details on how these layouts are applied to genomic tracks and matrices. This categorization facilitates understanding of how genomic features are displayed, whether through color, positional encodings, or parallel and orthogonal arrangement of sequence axes.
3. Task Taxonomy: The tasks are organized by the goals of visualization, ranging from lookup and browsing through specific features and loci, to multi-locus exploration and summarization across multiple feature sets.

Tools for Genomic Data Visualization

The paper reviews 83 tools categorized by their layout type (linear, circular, or space-filling), arrangement, and view configurations (views, scales, and foci). Notably, the authors discuss tools like the Integrative Genomics Viewer (IGV) and Circos, which utilize linear and circular layouts respectively, to handle non-interconnected and sparsely interconnected genomic features. For densely interconnected datasets, tools like Juicebox utilize orthogonal arrangements of linear layouts.

Implications and Future Opportunities

The development of increasingly sophisticated genomic data visualization tools has significant implications both theoretically and practically. As the field transitions towards integrating more diverse data types, including patient genomic data and dynamically changing genomic structures (such as 3D genome interaction data), there is a growing demand for visualization frameworks that efficiently handle multi-scale and heterogeneous datasets.

The paper identifies promising opportunities for advancing visualization techniques, particularly in managing 3D genome interactions and integrating genomics with broader biomedical data. This integration is expected to address evolving challenges such as secure data management and real-time analysis of genomic datasets, especially in precision medicine.

Conclusion

The paper delivers a comprehensive survey and taxonomy of genomic data visualization, establishing a foundational framework for future tool development and research in the domain. By dissecting current methodologies and challenging existing paradigms, it encourages the exploration of novel visualization techniques capable of handling the ever-increasing complexity and scale of genomic data. This work not only guides current genomic researchers but also provides a pivotal reference for future endeavors in the visualization community.