The revised general chapter, Chemometric methods applied to analytical data (5.21), has just been published in Supplement 11.1 of the European Pharmacopoeia (Ph. Eur.). The text was adopted by the Ph. Eur. Commission at its 172nd Session in March 2022.
Published for information, this general chapter provides an introduction to the use of chemometrics and data science techniques, together with recommendations for good practice and requirements for the processing of analytical data.
Data analysis is known to be an extremely dynamic field and the way algorithms are referenced and applied has undergone numerous changes since the general chapter was first published in 2015 (Supplement 8.7), prompting this revision.
In view of the latest developments, numerous sections of the chapter have been completely rewritten or updated and new sections/sub-sections have been introduced. This general revision includes:
- an update of section 1. General aspects with a review of the sub-sections on Pre-processing (1-2-2-6) and Assessment and validation of chemometric methods (1-3);
- two new sub-sections on Independent component analysis (2-2) and Decision trees and random forests (2-6);
- a general review of the sub-sections on Similarity measures (2-3), Clustering (2-5), Multiple linear regression (2-8), Principal component regression (2-9), Support vector machines for supervised classification (2-11) and Artificial neural networks (2-12);
- a new section, 3. Related application fields, with sub-sections on Chemometrics in chemical imaging (3-1), and Data fusion (3-2);
- an update of the Glossary and the Abbreviations.
The updated chapter now reflects the most current data science techniques in use and both the European Directorate for the Quality of Medicines & HealthCare and the Spectroscopy and Data Analysis working party responsible for its revision hope that it will be of continued assistance to users when analysing their data.
See also: