Background: Various laboratory-developed metabolomic methods lead to big challenges in inter-laboratory comparability and effective integration of diverse datasets.
Results: As part of the Quartet Project, we establish a publicly available suite of four metabolite reference materials derived from B lymphoblastoid cell lines from a family of parents and monozygotic twin daughters. We generate comprehensive LC-MS-based metabolomic data from the Quartet reference materials using targeted and untargeted strategies in different laboratories. The Quartet multi-sample-based signal-to-noise ratio enables objective assessment of the reliability of intra-batch and cross-batch metabolomics profiling in detecting intrinsic biological differences among the four groups of samples. Significant variations in the reliability of the metabolomics profiling are identified across laboratories. Importantly, ratio-based metabolomics profiling, by scaling the absolute values of a study sample relative to those of a common reference sample, enables cross-laboratory quantitative data integration. Thus, we construct the ratio-based high-confidence reference datasets between two reference samples, providing "ground truth" for inter-laboratory accuracy assessment, which enables objective evaluation of quantitative metabolomics profiling using various instruments and protocols.
Conclusions: Our study provides the community with rich resources and best practices for inter-laboratory proficiency tests and data integration, ensuring reliability of large-scale and longitudinal metabolomic studies.
Keywords: LC–MS; Metabolomics; Quality control; Reference material; Signal-to-noise ratio (SNR); Standardization.
© 2024. The Author(s).