Written in Python, this script contained a bug, discovered by researchers at the University of Hawaii, that could vary the results of calculations depending on the operating system used.
As Vice explains in an article published a few days ago, the software was used to make calculations of "chemical shift values for NMR, or nuclear magnetic resonance spectroscopy". In chemistry, this allows to know the molecular composition of a sample.
A "subtle" error
One of the researchers at the University of Hawaii discovered the bug by checking calculations for a project. The objective of this study was to find effective compounds against cancer.
Having observed inconsistencies, the group finally concluded that the results of calculations varied according to the OS used to launch this script.
Subsequently, they published a corrected version of the script. But because of the bug, already published scientific studies may have been skewed. "This simple problem in the original script challenges the conclusions of a significant number of articles on a wide range of topics in a way that can not be easily solved from published information because the operating system is rarely mentioned, "reads the new article. "Authors who have used these scripts should definitely check their results and any relevant conclusions using the modified scripts [..]"
For the moment, it is not clear exactly how many studies have been skewed. But according to Vice, which cites Nature Protocols data, the 2014 publication was consulted 1,900 times, and is cited in a further 158 studies. However, only studies that used the script to do calculations should be really affected.
Source link
https://www.presse-citron.net/une-simple-erreur-de-code-a-potentiellement-fausse-plus-dune-centaine-detudes-scientifiques/
Dmca
