Collected_stories

lost the backup

The external hard drive is VERY important to me as it contains 5 years of research data which are crucial for my PhD thesis !!! I think comments are unnecessary. Distributed under a CC-BY license, photo ©Peter Murray-Rust originally posted at https://blogs.ch.cam.ac.uk/pmr/2011/08/01/why-you-need-a-data-management-plan/ Collected by Julien Colomb, data manager

Continue reading

Exposing research data

In my work, I am the developer and administrator of a web based information system for research data applied in more than 30 research projects to share collected, brought or created data within the project consortium and with stakeholders. Even single researcher e.g. PhD students use the system to organize their own data. While most datasets can be downloaded without any extra request, few datasets can only be downloaded after the dataset owner approval.

Continue reading

paper retraction in absence of data

see copyrighted material here: http://science.sciencemag.org/content/357/6352/654.1 In brief, a paper was retracted from science because the raw data could not be presented. The computer containing the data has apparently been stolen and no backup existed. There are other similar stories on retractionwatch.com (https://retractionwatch.com/2016/02/23/we-are-living-in-hell-authors-retract-2nd-paper-due-to-missing-raw-data/, https://retractionwatch.com/2017/01/20/boom-headshot-disputed-video-game-paper-retracted/). It is not always clear if there were misconduct, but it seems the absence of data makes it impossible for researchers to prove they did not misconduct.

Continue reading

fly olfactory learning and lab meta-analysis

In 2007, I entered a lab doing mainly one experiment: olfactory learning assay in fruit fly. The output of the experiment is the number of flies choosing one or the other odor, recorded manually on spreadsheets. For the 14 people in the lab, they were approximately 20 different spreadsheets around. We managed to create one tidy template for all experiments in about 6 months. Interestingly, it took 2-3 of us to design the first draft, we then gathered feedback from the whole group, get a second version and so on.

Continue reading