Metlin metabolomics database

A matrix of molecular features characterized by mass to charge ratio (m/z) and retention time (RT) was generated using MassHunter Workstation Qualitative software version B07.00, MassHunter Profinder (version B.08.00), Mass Profiler Professional (MPP version 14.5) and Personal Compound Database Library (PCDL) (Agilent Technologies, CA, USA). Molecular features were extracted and binned/aligned using parameters as follows: peak height ≥ 10,000 counts, compound ion count threshold of two or more ions, compound alignment tolerances 0.00% + 0.15 min for RT and 20.00 mg kg⁻¹ ± 2.00 mDa for mass using Profinder. Molecular features which were present only in three samples out of five were included in the analysis. Compounds were tentatively identified by matching molecular entities with PCDL entries having similar accurate mass, RT, and mass spectra (generated from analytical standards) where possible and the METLIN metabolomics database (version B 07.00, Agilent Technologies, CA, USA) otherwise. All descriptive statistical analyses were performed using Statistix (Statistix software, version 4.1; Analytical Software, Tallahassee, FL, USA) and standard deviations were calculated and reported where possible.

A matrix of molecular features characterized by mass to charge ratio (m/z) and retention time (RT) was generated using Agilent MassHunter Workstation Qualitative software version B07.00, Agilent MassHunter Profinder (version B.08.00), Agilent Mass Profiler Professional (MPP version 14.5) and Agilent Personal Compound Database Library (PCDL) (Agilent Technologies, Santa Clara, CA, USA). The parameters for molecular extracts and peak binning/alignment using Profinder were as follows; peak height ≥ 10,000 counts, compound ion count threshold two or more ions, compound alignment tolerances were 0.00% + 0.15 minutes for RT and 20.00 ppm + 2.00 mDa for mass. The extracted ion chromatogram was smoothed with Gaussian smoothing before integration. The data files were converted to compound exchange files (.cef) format and visualized and analyzed in MPP using multivariate analysis including principal component analysis. Features present in a minimum of three out of five replicates were included for further analysis. Identification was performed comparing the generated database to the METLIN metabolomics database (version B 07.00, Agilent Technologies, Santa Clara, CA, USA) and confirmed using available standards based on accurate mass, retention time (RT) and mass spectra.