Visual analysis of time-series data on protein phosphorylation presents a particular challenge: bioinformatics tools currently available for visualising 'omics' data in time series have been developed primarily to study gene expression, and cannot easily be adopted to phosphorylation data, where a single protein typically has multiple phosphosites. In this study, we worked with an experimental research group that is applying very recent methods in high-throughput experimental proteomics to study the time course of protein phosphorylation events in human cells in vitro following stimulation by insulin, as part of a broader study of diabetes and obesity. We applied several existing visual analytics approaches with the goal of organising the data to facilitate new insight into underlying molecular processes. We developed a novel layout strategy called 'Minardo' that is loosely based on cell topology and ordered by time and causality. This layout utilises a frame of reference familiar to life scientists and helpful for organising and interpreting time-series data. This strategy proved to be useful, leading to new insights into the insulin response pathway. We are working on generalising the Minardo layout to accommodate similar datasets related to other signalling pathways, which should be straightforward.

This content is only available via PDF.
You do not currently have access to this content.