Reproducibility and transparency

Reproducibility and transparency in biology and medical sciences have been a hot topic since the early 2010s (Figure 9), with the beginning of what researchers call the “reproducibility crisis”. According to a survey conducted in 2016 by Nature on a pool of 1,576 scientists, in around 70% of the cases an attempt was made to reproduce someone else’s experiment, it was unsuccessful. By ‘reproducing’, we mean both in a ‘mechanical’ sense (e.g. starting from a given raw data file, following an exact set of analysis procedures to obtain an identical processed data file) and in a ‘biological’ sense (e.g. replicating the conclusion of a study by sourcing comparable samples and following similar wet-/dry-lab protocols).

Some steps that you can take to ensure the reproducibility and transparency of your experiment are:

• ensure that the design of your experiment includes appropriate controls and replicates
• use appropriate statistical tests and seek advice where necessary
• keep back-up copies of your raw data (request them from the core facility if needed)
• document all procedures and parameters (‘Materials and Methods’) with as much detail as possible
• submit your (well annotated) experiment to a public database for functional genomics experiments such as ArrayExpress or GEO

For more on ensuring reproducibility, review the page ‘Is it FAIR?’ in our online tutorial on data management.