TitleAutomating data analysis for two-dimensional gas chromatography/time-of-flight mass spectrometry non-targeted analysis of comparative samples.
Publication TypeJournal Article
Year of Publication2018
AuthorsTitaley, IA, O Ogba, M, Chibwe, L, Hoh, E, Cheong, PH-Y, Simonich, SLMassey
JournalJ Chromatogr A
Volume1541
Pagination57-62
Date Published2018 Mar 16
ISSN1873-3778
KeywordsBiodegradation, Environmental, Environmental Monitoring, Environmental Pollutants, Gas Chromatography-Mass Spectrometry, Polycyclic Aromatic Hydrocarbons, Software, Soil
Abstract

Non-targeted analysis of environmental samples, using comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC × GC/ToF-MS), poses significant data analysis challenges due to the large number of possible analytes. Non-targeted data analysis of complex mixtures is prone to human bias and is laborious, particularly for comparative environmental samples such as contaminated soil pre- and post-bioremediation. To address this research bottleneck, we developed OCTpy, a Python™ script that acts as a data reduction filter to automate GC × GC/ToF-MS data analysis from LECO ChromaTOF software and facilitates selection of analytes of interest based on peak area comparison between comparative samples. We used data from polycyclic aromatic hydrocarbon (PAH) contaminated soil, pre- and post-bioremediation, to assess the effectiveness of OCTpy in facilitating the selection of analytes that have formed or degraded following treatment. Using datasets from the soil extracts pre- and post-bioremediation, OCTpy selected, on average, 18% of the initial suggested analytes generated by the LECO ChromaTOF software Statistical Compare feature. Based on this list, 63-100% of the candidate analytes identified by a highly trained individual were also selected by OCTpy. This process was accomplished in several minutes per sample, whereas manual data analysis took several hours per sample. OCTpy automates the analysis of complex mixtures of comparative samples, reduces the potential for human error during heavy data handling and decreases data analysis time by at least tenfold.

DOI10.1016/j.chroma.2018.02.016
Alternate JournalJ Chromatogr A
PubMed ID29448996
PubMed Central IDPMC5909067
Grant ListP01 ES021921 / ES / NIEHS NIH HHS / United States
P30 ES000210 / ES / NIEHS NIH HHS / United States
P42 ES016465 / ES / NIEHS NIH HHS / United States
T32 ES007060 / ES / NIEHS NIH HHS / United States