Biotelligences poster has been awarded the Vifor Pharma special prize yesterday at the 2014 Swiss Pharma Science Day in Bern.

We thank the committee for this award and the organizers for the excellent meeting.

We look forward to the 2015 meeting.

Biomedical research is often image based; from the bright green cells generated by your transgene to the frenetic trace on your computer screen during an optogenetic experiment. Given this, we sought to generate visually meaningful word clouds, which are graphical representations of word frequency, from recent scientific publications. Our idea was to use these clouds to give visual information on the use of biostatistics by researchers.

Results paragraph: The obsession with P
We first used random text samples from Results sections and figure captions in 18 articles from 5 different journals (Nature, Neuron, Gastroenterology, Biological Psychiatry, British Journal of Pharmacology). The articles and journals were chosen randomly and a total of 2868 words were used. Words with very low occurring frequencies have automatically been excluded by the word generator.

This information clearly matches the results we collected from more than 300 articles in 25 periodicals indicating that: i) 98% of articles rely on p-values to conclude; ii) 84% use parametric tests (such as t-tests) despite detected violations of parametric assumptions; and iii) that 72% use SEM as an error bar, despite their well documented limitations (SEM do not inform the reader about variability but about precision of mean estimation, and the actual mean is not located within one SEM but within a 2-4 SEM-wide interval).

Statistics paragraph: Journal specificities
We are aware that not all statistical tests are cited in the Results sections, but rather in the Statistics paragraphs of Methods sections. Therefore, we also created the following word clouds, each corresponding to an individual journal, and made up of Statistics paragraphs of 4 articles (small/simple words not related to statistics were excluded).

Nature and Science:

Note the relative scarcity of words in the Science cloud, reflecting our difficulty to collect substantial statistical paragraphs. It is also remarkable that the names of statistical software are almost absent from the Science cloud. This points at a general lack of disclosure about statistical methods in Science, an observation supported by our quantitative study.

Neuron:

Note the importance gained by the Kolmogorov-Smirnov test, often used by electrophysiologists to analyse distributions of cell responses. Interestingly the word "one-tailed" is way larger than "two-tailed". Finally, it is worth mentioning that the arsenal of statistical software is larger in this cloud.

British Journal of Pharmacology:

In this cloud, the obsession for the "0.05" threshold and the domination of "SEM" over other error bars are visible again. However, the most visible difference concerns the popularity of ANOVA as compared to other tests, probably linked to multi-groups experimental designs as well as the occurrence of doses responses studies in pharmacology.

Conclusion
We were not aiming, in doing this, to carry out a scientific demonstration into the misuse of biostatistics. Our sample size of a few articles from a handful of journals would be too low to conduct proper descriptive statistics and no real quantitative investigation was performed. In addition, the clouds are only based on disclosed information making flaws by non-disclosure invisible. This later point is critical because a comprehensive and quantitative study that we are currently conducting on hundreds of articles shows that the absence of disclosure is a major flaw in biomedical publishing. Nevertheless, some intriguing features emerged from our clouds and many words highlighted here tend to be in line with the results we obtained in our quantitative studies. In particular the quest for "significant" "p-values" below the "0.05" threshold using parametric "Student's t-tests" or "ANOVA" is pervasive, even when parametric assumptions are not respected.

The Biotelligences team


All word clouds were generated on www.wordle.net