Proteomics and the p-value trap

Proteomics researchers are not generally well-trained in statistical methods—which often leads them to use the methods incorrectly, resulting in flawed conclusions. Researchers at IRIG are working on a solution to this problem, and their investigations led to four articles in international journals in 2018 and 2019. In particular, the researchers are looking at the risk of incorrectly interpreting the p-value, which indicates how significant a result is depending on the sample studied.

Proteomics is especially vulnerable to this type of error. Due to advances in spectrometry, proteomics researchers must manipulate growing volumes of data that are becoming more and more complicated to analyze. The IRIG team is raising proteomics researchers’ awareness of several issues, from understanding statistics terminology to designing protocols that make it easier to control sample analyses.

Contact: thomas.burger@cea.fr

Facebook

Twitter