Dr Francis Gacenga, QCIF’s eResearch Analyst at the University of Southern Queensland, has produced a guide to improving reproducibility using QCIF and ARDC resources.
Having reproducible research can enhance the demonstration of quality and increase the impact of one’s research.
Researchers can use QCIF and Australian Research Data Commons (ARDC) IT resources with assistance from QCIF eResearch Analysts to efficiently undertake, capture and organise the methods and results of their research, providing a pathway for efficient and transparent publication of reproducible results.
The below table is a quick, reproducible research start-up guide that can be used to identify available resources so researchers can publish their results, analysis, methods, data and scripts or code for others to test or reproduce them.
The guide provides a possible scenario based on available open source tools, QCIF and ARDC resources accessible at no cost to researchers at QCIF member universities.
The list of processes provided is not exhaustive and is meant as illustration of a generic research scenario that can be used by researchers to implement reproducibility.
Reproducibility is not easy but the research benefits, cost savings and value it creates are worth the effort.
Any queries, please contact Dr Francis Gacenga.
| RESEARCH PROJECT PHASE | KEY PROCESSES | QCIF and ARDC – TOOLS on NECTAR | SUPPORT | TRAINING |
| Planning and resourcing | Access for QCIF member researchers | Nectar* virtual machine (VM) QRIScloud data storage | QCIF Hacky Hours Nectar eHelp QCIF eResearch Analysts | QCIF-supported Software Carpentry sessions, workshops and lessons |
| Research initialisation | Packaging | Docker, Singularity, VirtualBox | ||
| Process documentation and quality control | Version Control | Git | ||
| Control – file management, editing transfer | Automation | Linux Shells, Make | ||
| Data analysis | Data analysis | R, RStudio, Python | ||
| Compute | Compute | QRIScompute | ||
| Presentation and Visualisation | Presentation and Visualisation | R, RStudio, LaTeX, Make | ||
| Publication of reproducible research artefacts | Research dissemination | QRIScloud, RDA, Zenodo, CodeMeta, JOSS | ||
| Project Data Storage – Archival | Preservation | QRIScloud, RDA |
* Nectar: National eResearch Collaboration Tools and Resources
References:
- http://ropensci.github.io/reproducibility-guide/sections/introduction/
- https://genome.sph.umich.edu/w/images/9/92/ReproducibleResearch-TechTalk.pdf
This article was first published on 16/05/2019.