Reproducible bioinformatics: why?

Firstly, you should be aware that much of current science is not reproducible. Even though most scientists (and indeed the general public) can predict using prediction-markets which studies will replicate, we still publish and cite rubbish. Small sample sizes can explain a considerable part of bad science within neuroscience. The predominance of small sample sizes has been facilitated by the philosophy that any study which shows p<0.05 is 'significant': however, as used in science currently, this approach treats evidence for negligibly small effect sizes as significant. Within bioinformatics, a related problem is that sloppy coding practises just render studies irreproducible.

A fundamental philosphy of the lab is thus that open science is reproducible science. You should familiarise yourself with:

• The reproducibility crisis

  • 5-HTTLPR: A Pointed Review
    • This is a no-holds barred review of how a tower of false results was built within genetics. Read this and consider how you can ensure you do not make the same mistakes.
    • https://slatestarcodex.com/2019/05/07/5-httlpr-a-pointed-review/
  • Border, Richard, et al. "No support for historical candidate gene or candidate gene-by-interaction hypotheses for major depression across multiple large samples." American Journal of Psychiatry 176.5 (2019): 376-387.
    • https://www.ncbi.nlm.nih.gov/pubmed/30845820
  • "1,500 scientists lift the lid on reproducibility", Nature
    • https://www.nature.com/news/1-500-scientists-lift-the-lid-on-reproducibility-1.19970

• Ongoing discussions about open science

  • "Training students for the Open Science future", https://www.nature.com/articles/s41562-019-0726-z

Read some selfish reasons for being reproducible here.