→ Slides doi:10.7490/f1000research.1112716.1
→ Video
Authors Boris Simovski, University of Oslo
Sveinung Gundersen, University of Oslo
Daniel Vodák, Oslo University Hospital
Abdulrahman Azab, University of Oslo
Diana Domanska, University of Oslo
Eivind Hovig, University of Oslo
Geir Kjetil Sandve, University of Oslo
AbstractGenomic investigations increasingly involve multiple genome-scale datasets (genomic tracks) representing diverse cell types and epigenetic factors. This raises the need for software tools that allow efficient management and appropriate statistical analysis of such data collections. The dataset lists in Galaxy represents a major advancement in this direction, allowing efficient management and per-dataset analysis of large numbers of datasets within a web-based system. A natural next step is to consider integrative analyses of collections of datasets. An example is to comparatively assess the co-occurrence of a dataset of disease-associated SNPs against a large number of datasets representing chromatin accessibility in diverse cell types. In addition to managing a large number of datasets (of chromatin accessibility), such an analysis requires efficient means of locating and compiling a collection of relevant datasets, as well as the provision of appropriate statistical measures.
We have developed GSuite Tools, a Galaxy-based analysis framework that offers a broad range of novel tools for managing and performing statistical analysis on collections of datasets (genomic tracks). The tools operate on GSuite files - an alternative to the standard Galaxy dataset lists that provides additional robustness and flexibility as needed for this range of tools. A prototype of GSuite Tools was presented at GCC2015. After a year of further focused work, GSuite Tools now has much broader capabilities, is based on a deeper statistical treatment of the problems, and is ready for practical use by the community. It is publicly available at: http://hyperbrowser.uio.no/gsuite
