Abstract

This dataset contains the complete experimental matrices and multivariate responses used for the optimization of Gas Chromatography-Mass Spectrometry (GC–MS) analytical conditions and extraction procedures in the simultaneous determination of Polychlorinated biphenyl (PCB), brominated flame retardants (BFR) and novel brominated flame retardants (NBFR) in human serum. Four CSV files are provided. DesignOfExperiment_RF_LOQ.csv reports the D-optimal and full-factorial experimental designs used to optimize retention factor (RF), chromatographic separation, and limit of quantification (LOQ), including variable levels for injection parameters, oven temperatures, and pulse conditions. DesignOfExperiment_extraction.csv contains the extraction-optimization DoE evaluating the effect of hexane and ethyl acetate volumes and SPE elution composition on analyte recovery and matrix effects. PrincipalComponentAnalysis_RF_LOQ.csv includes A/FWHM ratios for all target analytes, used to compute Principal Component Analysis (PCA) models that summarize global chromatographic performance during method refinement. PrincipalComponentAnalysis_extraction.csv provides recovery-based PCA input data for assessing extraction efficiency across pollutant classes. Together, these datasets support full reproducibility of the chemometric workflow applied in the associated study, enabling reconstruction of DoE models, PCA analyses, and optimization paths. Typical uses include method-development benchmarking, teaching of multivariate optimization strategies, and re-analysis of GC–MS development data under alternative criteria.