Review of contemporary chemometric strategies applied on preparing GC-MS data in forensic analysis

Gas chromatographic techniques, particularly gas chromatography-mass spectrometry (GC-MS) and gas chromatography-flame ionisation detector, are essential chemical instruments in analysing volatile or volatilisable forensic samples, e.g. ignitable liquids and drugs, attributed to their capability and sensitivity in separating a complex mixture of volatile organic compounds. However, interpreting the GC-MS profiles via visual inspection alone to solve a classification or discrimination problem can be notorious challenging due to the high dimensionality and presence of the inherent instrumental noise. Chemometric techniques are thereby often employed to facilitate data interpretation and provide probabilistic arguments on the outcome. Several reviews have been devoted to summarising and discussing the chemometric analysis of the high dimensionality forensic data. However, none of the works places attention on chemometric strategies applied to preparing the GC-MS data but on the general descriptions and applications of a wide range of data preprocessing, peak alignment algorithms and modelling techniques in a selected discipline of knowledge. Therefore, this work aims to critically discuss the chemometric strategies applied to prepare the GC-MS data for forensic analysis. Several contemporary strategies applied in solving forensic problems were briefly described based on 42 selected articles. Then, open questions that remained unanswered were discussed and highlighted. This article aims to complement other reviews addressing general GC data analysis and forensic data interpretation.

Automated summary

This article discusses the chemometric strategies applied to prepare GC-MS data for forensic analysis, describing several contemporary strategies based on 42 selected articles. It aims to complement other reviews addressing general GC data analysis and forensic data interpretation, while highlighting unanswered open questions.