Storing provenance data of genome project workflows using graph database
- 1 November 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 16-22
Abstract
Many scientific experiments are designed as computational workflows in bioinformatics. However, the amount of data generated increases at every phase of each execution, hindering the identification of the source and the transformation of data. Therefore, it has become necessary to create new tools to store data provenance, mainly which resources and parameters were used to generate the results, among other information, to validate and publish the experiment. In this paper, we propose to use graph database to store data provenance using the PROV-DM model of bioinformatics workflows. To validate the model, we developed a simulator that worked as a logbook to capture data provenance. A workflow with real genomic data showed that very little additional data should be stored, which means that our provenance model can be easily included in genome projects.Keywords
This publication has 9 references indexed in Scilit:
- Automatic capture of provenance data in genome project workflows2013
- A Discussion on the Design of Graph Database BenchmarksLecture Notes in Computer Science, 2011
- A comparison of a graph database and a relational database2010
- Workshop on theory and practice of provenance event reportACM SIGMOD Record, 2009
- A Strategy for Provenance Gathering in Distributed Scientific Workflows2009
- Survey of graph database modelsACM Computing Surveys, 2008
- Modern Genome Annotation2008
- Why and Where: A Characterization of Data Provenance2001
- Genome annotation: from sequence to biologyNature Reviews Genetics, 2001