L’open science en pratique : un article modèle pour vos futures publications !

Vous voulez découvrir un bel exemple de la mise en application des principes de l’open science : reproductibilité, utilisation de logiciels open source, open data et publications en open access avec licence Creative Commons (CC-BY), etc. ? 

Le combo gagnant se trouve dans un des articles publié par Guillaume Dumas (co-fondateur de HackYourPhD) et ses collègues Yang-Min Kim Jean-Baptiste Poline. L’article “Experimenting with reproducibility: a case study of robustness in bioinformatics” (Juillet 2018) en mode meta vous explique en détail les principes mêmes de la reproductibilité tout en abordant les limites actuelles de sa mise en oeuvre et comment y pallier.


A lire et à partager !



Kim, Yang-Min, Jean-Baptiste Poline, et Guillaume Dumas. « Experimenting with Reproducibility: A Case Study of Robustness in Bioinformatics ». GigaScience 7, no 7 (1 juillet 2018). https://doi.org/10.1093/gigascience/giy077.


Reproducibility has been shown to be limited in many scientific fields. This question is a fundamental tenet of scientific activity, but the related issues of reusability of scientific data are poorly documented. Here, we present a case study of our difficulties in reproducing a published bioinformatics method even though code and data were available. First, we tried to re-run the analysis with the code and data provided by the authors. Second, we reimplemented the whole method in a Python package to avoid dependency on a MATLAB license and ease the execution of the code on a high-performance computing cluster. Third, we assessed reusability of our reimplementation and the quality of our documentation, testing how easy it would be to start from our implementation to reproduce the results. In a second section, we propose solutions from this case study and other observations to improve reproducibility and research efficiency at the individual and collective levels.While finalizing our code, we created case-specific documentation and tutorials for the associated Python package StratiPy. Readers are invited to experiment with our reproducibility case study by generating the two confusion matrices (see more in section “Robustness: from MATLAB to Python, language and organization”). Here, we propose two options: a step-by-step process to follow in a Jupyter/IPython notebook or a Docker container ready to be built and run.



Leave a Reply

Your email address will not be published. Required fields are marked *