We use the principles of information visualization to guide the design of systems to best meet the needs of
specific targets group of users, namely biologists who have different tasks involving the visual exploration of
biological networks. For many biologists who explore networks of interacting proteins and genes, the topological
structure of these node-link graphs is only one part of the story. The Cerebral system supports graph layout in
a style inspired by hand-drawn pathway diagrams, where location of the proteins within the cell constrains the
location within the drawing, and functional groups of proteins are visually apparent as clusters. It also supports
exploration of expression data using linked views, to show these multiple attributes at each node in the graph.
The Pathline system attacks the problem of visually encoding the biologically interesting relationships between
multiple pathways, multiple genes, and multiple species. We propose new methods based on the principle that
perception of spatial position is the most accurate visual channel for all data types. The curvemap view is an
alternative to heatmaps, and linearized pathways support the comparison of quantitative display as a primary
task while showing topological information at a secondary level.