PhD Retreat: Open Science and reproducibility

Unlocking reproducibility for your own research

19.03.24

Henrik Finsberg

Plan this workshop

• What is reproducibility, and why do we need it?
• Organizing your projects (version control)
• Recording dependencies and environment
• Recording computational steps (exercises)
  • Testing
  • Automation with GitHub workflows
  • Docker
• Example papers and concluding remarks

Material from the workshop

• https://github.com/scientificcomputing/phd-retreat-190324
• https://scientificcomputing.github.io/phd-retreat-190324

Based on material from

• https://github.com/scientificcomputing/seminar-23-11-2023
• https://coderefinery.github.io/reproducible-research/
• https://the-turing-way.netlify.app/reproducible-research/reproducible-research

About me

• 2014: Master in applied mathematics from NTNU, Trondheim

• 2014 - 2017: PhD: Computational Physiology, Simula

• 2017 - 2021: Research Engineer, Computational Physiology, Simula

• 2021 - present: Senior Research Engineer, Computational Physiology, Simula

• 2018 - present: Lecturing at UiO: IN1910 - programming for scientific applications

• I work on developing research software for cardiac applications

• I am a tools geek, and like to try out new software tools that can help me in my work

Image is used under a CC-BY 4.0 license. https://twitter.com/HeidiBaya/status/1579385587865649153.

Scenario 1

• You have submitted your first paper to a journal
• One of the reviewers asks for new figures and some additional analysis
• You rerun the analysis but you are unable to recreate the figures from the initial submission

• Alternative version: A new PhD student starts, and will be building upon previous work done by you

Baker, M. 1,500 scientists lift the lid on reproducibility. Nature 533, 452–454 (2016). https://doi.org/10.1038/533452a.

A reproducible article needs a good foundation

Steeves, Vicky (2017) in “Reproducibility Librarianship,” Collaborative Librarianship: Vol. 9: Iss. 2, Article 4. Available at: https://digitalcommons.du.edu/collaborativelibrarianship/vol9/iss2/4]

Discussion in groups (department-wise)

• Did you find any issues while trying to reproduce the results article that was handed out?
• What was missing?

What is reproducibility?

• A result is reproducible when the same analysis steps performed on the same dataset consistently produces the same answer.
• Same data + Same code -> Same results

Taken from https://the-turing-way.netlify.app/reproducible-research/overview/overview-definitions under a CC-BY 4.0 license

Advantage of working reproducible?

Advantage of working reproducible?

Taken from https://the-turing-way.netlify.app/reproducible-research/overview/overview-benefit under a CC-BY 4.0 license

Barriers to reproducibility

• Why is it so hard to make research reproducible?

Used under a CC-BY 4.0 license. DOI: 10.5281/zenodo.2669547.#

Next up: Organizing your projects

https://scientificcomputing.github.io/phd-retreat-190324/organizing-slides