![]() ![]() We show in a use case that our methods and the Pathfinder tool can help answer important questions about gene regulatory networks, when analyzing the networks in the context of rich experimental measurements. A demo version of Pathfinder is available at. These immediate results can be used to evaluate and refine queries, even before a complete and comprehensive answer is available. We employ strategies to deliver fast and progressive results, i.e., paths are added to the visualization as soon as they are found. We pay particular attention to scalability, so that we can handle tens of thousands of nodes and edges interactively. ![]() We realize and test our technique in a prototypical implementation. We also have developed a list of requirements for path analysis that we use to justify and evaluate our design and analyze related techniques. We introduce methods to (a) interactively query for paths and dynamically refine queries, (b) visualize the resulting paths and their relationships, (c) investigate the attributes associated with the paths, and (d) rank and compare paths. ![]() Our primary contribution is Pathfinder, a technique for the visual analysis of paths queried from large and multivariate networks. In this paper, we introduce methods to comprehensively address path analysis tasks. In larger networks, however, queries are essential to enable these tasks. For small networks, these path analysis tasks can be solved by visually finding paths, for example, in a node-link layout. Learning about how two suspects are connected in a criminal case or understanding why two genes are co-regulated are examples of important domain tasks that can be abstracted to path analysis tasks. Many tasks on such large graphs cannot be addressed by showing all nodes and links, even if a layout could be drawn.Īn important class of tasks for graph analysis is concerned with paths. As more data is collected and the graphs become bigger, scalable methods to extract knowledge and reason about them become more important. However, the visualization and analysis of these rich attributes present additional challenges, as there is a trade-off between optimizing a layout for conveying topology and attributes. In many cases, only the combined analysis of attributes and topology can lead to meaningful insights. Graphs are also increasingly associated with rich node and edge attributes. Also, interaction, for example, through queries, plays a critical role in tackling scalability problems. ![]() The analysis of graphs of nontrivial size depends on a combination of algorithmic, statistical, and visual approaches. It is now common to encounter networks that cannot be sensibly drawn due to both computational and perceptual constraints. Graph analysis and visualization have always been important for scientific discovery and decision-making, but their role and ubiquity have increased in the last decade. Graphs capture relationships between items, for example, friendships between people in social networks, interactions of genes in biological networks, or researchers coauthoring scientific papers in collaboration networks. We demonstrate Pathfinder's fitness for use in scenarios with data from a coauthor network and biological pathways. Pathfinder is designed to scale to graphs with tens of thousands of nodes and edges by employing strategies such as incremental query results. The paths can be ranked based on topological properties, such as path length or average node degree, and scores derived from attribute data. For the paths in the list, we display rich attribute data associated with nodes and edges, and the node-link diagram provides topological context. The resulting set of paths is visualized in both a ranked list and as a node-link diagram. We introduce Pathfinder ( Figure 1), a technique that provides visual methods to query paths, while considering various constraints. We show that by focusing on paths, we can address the scalability problem of multivariate graph visualization, equipping analysts with a powerful tool to explore large graphs. In this paper, we present visual analysis solutions dedicated to path-related tasks in large and highly multivariate graphs. These attributes are often critical in judging paths, but directly visualizing attributes in a graph layout exacerbates the scalability problem. Also, many networks are multivariate, i.e., contain rich attribute sets associated with the nodes and edges. Unfortunately, graph layouts often do not scale to the size of many real world networks. Typically, path-related tasks are performed in node-link layouts. The analysis of paths in graphs is highly relevant in many domains. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |