Graduate student from the University of Hawaii a problem in the Python script used for calculations , which defines the chemical structure of the substance being studied during spectral analysis of signals by means of . During the verification of the research results of one of his professors, the graduate student noticed that when the script was run on different operating systems with the same data set, the output produced different results.
For example, when running on macOS 10.14 and Ubuntu 16.04 for the tested data set, the script an incorrect value of 172.4 instead of 173.2. The script includes about 1000 lines of code and has been used by chemists since 2014. An examination of the code revealed that the output of the incorrect value differences in file sorting in different operating systems. The authors of the script believed that the function “” always returns files sorted by name, while the documentation for glob indicates that the order of output is not guaranteed. The fix involved adding list_of_files.sort() after the call to glob().
The identified problem has called into question the accuracy of over 100 publications in chemistry, the conclusions of which were made based on the calculated chemical shifts from the script. The exact number of studies that used the script is unknown, but the publication with its code has been cited in 158 works. The authors of these works have been advised to assess the correctness of the script's operation on the operating systems applied for calculations and to re-evaluate to ensure the correctness of the computed values. The incident is an excellent example of how not only the quality of experiment execution but also the correctness of data processing in programs that have previously been widely used can affect the final result.
The graduate student from the University of Hawaii discovered a problem in.
Source: opennet.ru
