Quantification precision and accuracy

Precision and accuracy were calculated on the mix of 54 deuterated standards designed for the Lipidyzer platform (SCIEX, cat#: 5040156, LPISTDKIT-101), which were spiked in lipid extracts from a reference human plasma sample (SCIEX, cat#: 4386703). Internal standards covered 10 main lipid classes found in human plasma with different final concentrations reflecting their physiological concentrations. LWM recommended concentrations spanning 3 orders of magnitude were used to determine precisions (Supplementary Table ²). Lipid extracts were prepared using methanol, methyl tert-butyl ether (MTBE) and water as previously described¹⁷ and stored at −20 °C in 10 mM ammonium acetate in 90/10 methanol/toluene before analysis. Intra-day precision was calculated as coefficient of variation (CV) using five replicate injections on the same day. Inter-day variability was determined by calculating the CV between the median of five replicate injections across three consecutive days. Precisions were calculated after median normalization for LC-MS data. Calibration curves were constructed for each IS using a 7-point dilution series spanning more than 2 orders of magnitude around the LWM recommended spike-in concentration (see Supplementary Table ² for details). The determination coefficient (r²) was calculated for each IS calibration curve (Supplementary Figures ² and ³). Accuracy was determined for each IS at the LWM recommended concentration by calculating the percentage of deviation of the calculated concentration relative to the actual concentration as follows: