An Overview + Detail Layout for Visualizing Compound Graphs

TL;DR

This paper tackles visualizing compound graphs with deep, tree-like nesting by introducing an overview+detail layout that preserves high-level structure while exposing inner group details. It achieves this with ports to route inter-group edges and a tailored Orthogonal Reingold-Tilford variant to place expanded subgraphs near their collapsed counterparts, enabling simultaneous assessment of multiple hierarchy levels. The authors demonstrate the approach through case studies in data-flow analysis and computational model extraction, highlighting debugging and comprehension benefits. The work discusses appropriate use cases, limitations, and avenues for future enhancement, emphasizing practical impact for domain experts handling nested, multi-level networks.

Abstract

Compound graphs are networks in which vertices can be grouped into larger subsets, with these subsets capable of further grouping, resulting in a nesting that can be many levels deep. In several applications, including biological workflows, chemical equations, and computational data flow analysis, these graphs often exhibit a tree-like nesting structure, where sibling clusters are disjoint. Common compound graph layouts prioritize the lowest level of the grouping, down to the individual ungrouped vertices, which can make the higher level grouped structures more difficult to discern, especially in deeply nested networks. Leveraging the additional structure of the tree-like nesting, we contribute an overview+detail layout for this class of compound graphs that preserves the saliency of the higher level network structure when groups are expanded to show internal nested structure. Our layout draws inner structures adjacent to their parents, using a modified tree layout to place substructures. We describe our algorithm and then present case studies demonstrating the layout's utility to a domain expert working on data flow analysis. Finally, we discuss network parameters and analysis situations in which our layout is well suited.

Figures (7)

• Figure 1: An overview+detail layout (a) vs. focus+context layout (b) of the same compound graph. The graph depicts the data flow of a small Python program. In our case study, a domain expert looks at compound networks of larger, more complicated programs.
• Figure 2: Compared to (a), we explicitly incorporate ports into the design in (b) to increase the clarity of links through compound graphs and isolate the the layout within compound nodes.
• Figure 3: (Top) The Reingold-Tilford algorithm can produce poor results when the compound node is oblong in the direction of tree growth. Even growing the layout vertically, the internal view would be placed in the middle of the oblong compound node rather than near its counterpart. (Bottom) The ideal layout would place expanded nodes near their collapsed counterparts.
• Figure 4: An illustration of our proposed variant of RT algorithm. The input data is shown in both our layout and a simplified tree view. Then in each step, we follow the RT bottom-up placement. We frist place group parents with respect to expanded children based on the position of their corresponding internal nodes, and then make separation passes in both directions of tree expansion.
• Figure 5: Duplicate structures within one group are drawn with edges from the collapsed copies to the expanded form. In this application, we choose lightweight dashed edges to depict this relationship. Users of our layout can choose to suppress these edges and/or the expansion of duplicate siblings as layout parameters..