Getting poster data...
Xin-Yi Chua, James M. Hogan, Daniel Johnson (Queensland University of Technology, Brisbane, Australia)Network structures are pervasive in bioinformatics, notably in the representation of complex regulatory relationships, such as those between transcription factors and their target genes. The better we understand the system, the more dense and impenetrable the graph may become when visualised. The modern abundance of molecular sequencing data enhances understanding, but it also offers the chance to compare networks across many species and strains, helping to identify the variations which mediate functional differences. Such comparisons are often futile if full regulatory networks are shown, and one approach to managing this complexity is to use iconic, structurally similar subnetworks to facilitate grouping and further analysis. Yet these analyses may involve many hundreds of network icons, raising important questions about our cognitive limitations as visualisations scale in the number of individual networks, the complexity of each network, the resolution of the image and the size of the display. This study will examine how judgments of network similarity vary with scale with respect to these design factors.