A Suite of Ontologies
for Preserving Workflow Research Objects

This web page describes the Research Object and all the resources associated to the paper with the same title currently submitted at the Journal of Web Semantics. The paper describes four ontologies for representing workflows in Research Objects, and includes examples and motivation scenarios.

Abstract

Scientific workflows have, in recent years, gained momentum in modern sciences as one of the popular means for specifying and automating data-driven in silico experiments. The value of scientific workflows also lies in their potential reusability. A workflow provides an explicit specification of an experiment. Once shared, workflow specifications become useful building blocks to be combined or modified for developing new experiments. However, storing specifications is not sufficient for preserving workflows. Scientists need to have an understanding of the workflow and how it may be used and repurposed for their needs. For this purpose, additional resources such as annotations describing the workflow, datasets used and produced by the workflow, and provenance traces recording workflow executions, may be needed to preserve workflows. In this article, we present a suite of ontologies that can be used for preserving scientific workflows. The design of the ontologies was guided by requirements that were raised by scientists involved in the wf4ever project, as well as by our experience in dealing with workflow decay in previous work. The ontologies developed make use of and extend existing well known ontologies, namely the Object Reuse and Exchange (ORE) vocabulary, the Annotation Ontology (AO) and the W3C PROV ontology (PROVO). We illustrate how the ontologies can be utilized using a real-world scenario, in which scientists created a Workflow Research Object for an investigation on the Huntington's disease. We also present the tools we developed for managing Workflow Research Objects.

Inputs and examples of the analysis

Motivation scenario: the Huntington's Disease
Huntington's disease (HD) is the most common inherited neurodegenerative disorder in Europe, affecting 1 out of 10000 people. Although the genetic mutation that causes HD was identified 20 years ago, the downstream molecular mechanisms leading to the HD phenotype are still poorly understood. Relating alterations in gene expression to epigenetic information might shed light on the disease aetiology. In the context of the Wf4ever project, a team of scientists from the Leiden University Medical Centre (HG-LUMC) investigates this hypothesis using workflows.

Figure 1: A sketch depicting the main steps that the bioinformatician followed for manipulating and analyzing datasets, and the workflows that were used in each step

Reused Workflows and Research Objects Figure 1 illustrates the main steps that the bioinformaticians followed in their in silico analysis. The figure sketches out the datasets used as inputs in the analysis, the steps that compose the analysis, and the workflows implementing the steps of the analysis. The bioinformaticians implemented two types of workflows to serve their in silico analysis. The first two workflows, getConceptsSuggestionsFromTerm.t2.flow and match_two_gene_lists_prioritize_gene_list.t2flow are used for gene prioritization, and are executed in sequence.

The third workflow is used to annotate the genes, that are retrieved from five datasets. Once the workflows have been executed, the bioinformaticians analyzed the results to reach their conclusions. The analysis of the workflow results is not illustrated in the sketch depicted in Figure 1 for simplicity.

A research object with executions of the third workflow can be accessed here

Paper results and resources

Our results can be summarized as four main ontologies

1. Wfdesc, ontoloy used for specifying workflow descriptions. Here is an example modeling the third workflow of the Research Object.
2. Wfprov, ontology used to describe workflow executions. Here is an example modeling an execution of the third workflow included in the Research Object.
3. WRO, ontology used to describe general Research Objects and their resources.Here is an example.
4. ROEvo, ontology for describing the evolution of Research Objects.Here is an example.

About the authors

Acknowledgements