Evaluation of Nontarget Long-Term LC-HRMS Time Series Data Using Multivariate Statistical Approaches

Area of Science:

• Analytical Chemistry
• Environmental Science
• Chemometrics

Background:

• Liquid chromatography-high-resolution mass spectrometry (LC-HRMS) is increasingly used for nontarget screening (NTS).
• NTS generates vast datasets, complicating data interpretation and temporal monitoring.
• Identifying unknown trace organic compounds (TrOCs) in complex matrices like industrial wastewater is challenging.

Purpose of the Study:

• To develop and validate a chemometric prioritization method for handling time-series NTS data.
• To reduce raw data complexity and facilitate the identification of unknown contaminants in industrial wastewater.
• To demonstrate the method's efficacy in revealing temporal trends of TrOCs.

Main Methods:

• Utilized a five-month time series of industrial wastewater samples analyzed by LC-qTOF-MS.
• Applied nontarget screening (NTS) workflows including peak detection, alignment, grouping, and blank subtraction.
• Employed featurewise principal component analysis (PCA) and groupwise PCA (GPCA) for time trend detection and feature prioritization.

Main Results:

• Processed 3303 features from wastewater treatment plant (WWTP) influent samples.
• Reduced features to 130 relevant time trends associated with TrOCs.
• Successfully identified *N*-methylpyrrolidone as a prioritized nontarget pollutant.

Conclusions:

• The developed chemometric strategies effectively prioritize unknown contaminants in complex time-series data.
• The method significantly reduces data complexity, enabling efficient identification of TrOCs.
• This approach is applicable to industrial wastewater analysis and other scientific fields requiring time trend exploration.