Networks are widely used throughout almost every field of study and have been applied to an extraordinarily wide range of problems, such as the structure of language, social group analysis, military communication networks designed for robustness in the event of nuclear attack, co-publication graphs, or who has acted with whom in Hollywood movies.
Within the biosciences, proteins interaction networks, species trees, metabolic networks, gene regulatory networks and many other areas of bioinformatic interest have benefited from network analysis.
The essence of a network is to model pairwise relations between objects by defining a set of nodes, and edges that connect the nodes and represent “being related”.
With such a diverse range of applications, methods of visualisation or drawing of networks to organise, make sense of, and generate hypotheses to test are of great interest.
A well-drawn network simplifies complexity, and can often be the difference between a mass of unstructured data and a new understanding of previously hidden relationships. It will allow you to easily answer questions, such as what are the most important parts of the network? Are there distinct components? How closely related are particular components? What are the most/least highly connected components? How fragile/breakable is the network? And from these observations, find patterns, and make inferences and predictions.
In this hands-on workshop, an introduction will be given to how networks or graphs are defined and methods to visualise them. The primary tool will be Cytoscape, a free, powerful multi-platform application for visualisation and analysis of networks.
Workshop topics covered will include:
- simple formats, including Excel and text documents, for describing and creating networks
- easy interactive network creation in Cytoscape
- different methods of automated layout so that your network is comprehensible and visually pleasing
- changing the style of components of the network by mapping arbitrary attributes to colour, size, shape, or images, or charts
- the Cytoscape App manager for extending Cytoscape’s capabilities
- Cytoscape’s Network Analyser for easy statistics generation and analysis of your networks
- exporting high resolution images for publication, and
- getting help when you are stuck
- If there is time, we may also cover sub-sampling, aggregation and pruning as approaches to making sense of “hairballs”.
Examples used will include the classic sociograms of Moreno—describing the changing social relations between schoolchildren, visualising co-publication networks generated from Scopus literature searches, and protein interaction networks.
At the end of this workshop, you will know how to:
- easily describe networks in simple file formats
- import and visualise these into Cytoscape
- highlight and find information in your networks by mapping node and edge attributes to colours, shapes and sizes
- try different layout methods to find the optimum one for your problem, and
- generate publication-quality figures.