Data cleaning and QAQC pipeline for targeted PFAS LCMS/MS from Sciex

Abstract: Per-and polyfluoroalkyl substances (PFAS) are persistent and ubiquitous environmental contaminants with harmful effects on human health. They are contained in everyday life products such as textiles and water/grease resistant coatings along with firefighting foams. Long-range transport of PFAS around the world is facilitated by hydrologic and atmospheric processes and bioaccumulation in animals. One central goal of the Superfund research center, STEEP, is the detection of PFAS in environmental matrices such as water, sediment, and biospecimens (i.e., liver and serum from mice and humans). Passive sampling techniques are frequently employed to aid in the detection of PFAS. Detection and quantification of 60 targeted PFAS compounds are measured by liquid-chromatography tandem high-resolution mass spectrometry (LC-HRMS/MS). Analysis is conducted with SCIEX instrumentation and software. The raw data product of SCIEX is a cumbersome table that needs to be further analyzed using a harmonized approach. Jupyter notebooks have been developed to automate and standardize data processing of results from targeted LC-HRMS/MS analysis. By integrating code, data, and narrative in a single, shareable document, they help ensure that analytical workflows are transparent, reproducible, and consistently formatted. The developed routine combines raw data from multiple sets of samples (batches) and runs (different sets of targeted compounds), and flags values failing the quality control criteria. Quality control criteria incorporated in the routine have been adopted from guidelines of the Environmental Protection Agency (EPA, 1633A) and compared to guidelines of the Department of Food and Drug Administration (FDA, Q2(R1)). To maintain the necessary flexibility for research and work across disciplines, certain criteria and calculation methods can be adjusted by researchers, and threshold values can be set as input parameters. The output of the routine consists of a lon-format table comprising the most relevant information in open format (.csv), as well as a spreadsheet. The spreadsheet is used to guide researchers through quality control and further investigate details of the LC-HRMS/MS measurement.