ABSTRACT: Lipids were quantified in human plasma from the 1:3 matched TEDDY case-control subjects (Lee et al., 2013). Blood plasma was extracted following the methyl-tert-butyl ether (MTBE) extraction protocols. The choice of internal standards and chromatographic conditions was optimized by using toluene in the reconstitution solvent mixture to ensure that very lipophilic components like cholesteryl esters (CEs) and triacylglycerols (TAGs) are efficiently transferred to the ultrahigh-pressure liquid chromatography (UHPLC) column in the injection process. Lipids were analyzed by charged surface hybrid column electrospray ionization quadrupole time of flight tandem mass spectrometer (CSH-ESI QTOF MS/MS). The analytical ultra-high-pressure liquid chromatography (UHPLC) column is protected by a short guard column which was replaced after 400 injections while the UHPLC column was replaced after 1,200 serum (or plasma) extract injections. The sequence of column replacements were evaluated to ensure no detrimental effects were detected with respect to peak shapes, absolute or relative lipid retention times or reproducibility of quantifications. Automatic valve switching was used after each injection to reduce sample carryover for highly lipophilic compounds. This valve switching employed a dual solvent wash, first with a water/acetonitrile mixture (1:1, v/v) and subsequently with a 100% isopropanol wash. LC-BinBase was used as an untargeted approach for annotating chromatographic peaks and spectra against a dynamically built library of compounds. The quantified raw dataset was normalized by the SERRF bioinformatics pipeline [1]. In the LC-QTOF data, samples were removed before normalization if they failed the laboratory’s QC standards or were missing data for more than half of the compounds. The SERRF normalized data have been made available for identified compounds. Results data for unidentified compounds are available un-normalized. An explanation of the study design variables are explained in detail in a data dictionary provided in the raw data download section. References: 1) Lee HS, Burkhardt B, McLeod W, Smith S, Eberhard C, Lynch K, Hadley D, Rewers M, Simell O, She JX, Hagopian W, Lernmark A, Akolkar B, Ziegler AG, Krischer J, and the TEDDY Study Group: Biomarker discovery study design for type 1 diabetes in The Environmental Determinants of Diabetes in the Young (TEDDY) study. Diabetes/Metabolism Research and Reviews. Epub 2013 December 15. doi: 10.1002/dmrr.2510 (PubMed ID: 24339168). 2) Fan S, Kind T, Cajka T, Hazen SL, Tang WHW, Kaddurah-Daouk R, Irvin MR, Arnett DK, Barupal DK, Fiehn O: Systematic Error Removal Using Random Forest for Normalizing Large-Scale Untargeted Lipidomics Data. Anal Chem 2019, 91(5):3590-3596.
Jeffrey Krischer 
