Mass hunter profinder b 06

Metabolite feature extraction was performed with an Agilent Mass Hunter Profinder B.06.00 in pooled plasma samples from the MPP. Metabolite features that were present in ≥80% of the pooled samples (N = 1908) were extracted from analytical samples and quality control samples. Normalization was performed via metabolite measurements in the quality control samples. A low-order nonlinear locally estimated smoothing function was fitted to the measurements of each metabolite feature in the quality control samples. These functions were used to create correction curves, to which each metabolite feature measurement in the analytical samples were normalized (17 (link)).
Only metabolite features that had a coefficient of variation <20% in the repeated injections of quality control samples and were measured in >90% of the participants in MPP were selected for statistical analysis (N = 1025) to ensure high repeatability of studied metabolite features. Metabolites that were previously investigated in the same study material, via a targeted metabolomics method, were excluded from the analysis (4 (link)). The study workflow is presented in an online repository (14 (link)).

For semi-targeted analysis, LC-MS data were processed using MassHunter Profinder B.06.00 software (Agilent Technologies, Santa Clara, CA, USA). The Batch Targeted Analysis option was used with the compounds library, built based on previously published results [13 (link)]. The library included both compound formula and molecular mass. During data inspection the quality of the MS spectra was verified as well as the comparison of metabolites retention times with those obtained previously was made. Visual inspection of the mass spectra allowed the elimination of false-positive results in the created data matrix.
For GC-MS data the compounds were selected from data matrix prepared for untargeted analysis.
Selected signal intensities (from LC-MS and GC-MS analyses) were normalized in three ways: with the use of SG values, creatinine (CR) concentration (both measured with the use of portable urine analyzer: Urit 31, URIT Medical Electronic Co., Guangxi, P.R. China) or by using the abundance of creatinine MS signal obtained during analyses. A comparison of results of statistical analysis performed on those 3 datasets can show whether results will differ depending on the selected method of normalization.