A Linked Data Approach to Sharing Workflows and Workflow Results

Abstract

Abstract.A bioinformatics analysis pipeline is often highly elaborate, due to the inherent complexity of biological systems and the variety and size of laboratory publications would be highly beneficial to bioinformatics, for evaluating evidence and examining data across related experiments, while introducing the potential to find associated resources and integrate them as data and services. We present initial steps towards preserving bioinformatics design of a data analysis pipeline, and RDF to link the workflow, its component services, run-time provenance, and a personalized biological interpretation of the results. An example shows the reproduction of the unique graph of an analysis procedure, its results, provenance, and personal interpretation of a text mining experiment. It links data from Taverna, myExperiment.org, BioCatalogue.org, and ConceptWiki.org. The approach is relat -unobtrusive to bioinformatics users.

Machines, methods and music: On the evolution of e-Research

“…This is made increasingly possible by the improving circumstances of reuse in the digital research ecosystem and particularly by the adoption of Linked Data 4 . As well as publishing Linked Data workflows, myExperiment itself is part of that fabric, nicely exemplified in Roos and Marshall's proof-of-concept mashup in which they demonstrate an assembly of resources to answer a research question and in doing so make a point about the future methods section of papers [20].…”