Submitted by S.E. Morgan on Tue, 23/04/2019 - 15:39
The biomedical sciences are awash with high dimensional data describing the activity or abundance of genes, proteins and biochemicals. Graphia has been designed to turn such data into correlation networks and to render the often millions of nodes and edges in 3D space, where data structure can be explored and integrated. Due to the size and diversity of networks, Graphia has been specifically designed for speed, scale and agility.
In the context of the biomedical sciences, application areas for correlation analysis include: transcriptomics (RNA-seq, single cell), proteomics, metabolomics, multi-parametric FACS, genome diversity etc., but also the visualisation and analysis of protein interaction data, phylogenetic trees or any relationship matrix. But equally data or networks can be derived from any source.
Graphia runs on all desktop environments and includes a range of built-in graph analytics, e.g. PageRank, MCL clustering, enrichment analysis etc. and a wide range of options for graph transformation, i.e. filtering of nodes and edges based on attribute information. It also offers dynamic graph layout so any change to the graph structure is immediately reflected in the layout. You can also call up web pages directly from the node or link to a database of your choice. In terms of visualisation, then there is an advanced range of options to colour/size nodes and edges based on attribute data (imported or calculated within the tool).
Import of graph-based data is based on standard formats (.gml, .graphml) or more simple text-based formats (.txt, .csv).
Graphia is made produced by Kajeka, a University of Edinburgh spin-out company and is available for a 30-day free trial https://kajeka.com/graphia/