High-Resolution Filtering for Improved Small Molecule Identification via GC/MS

Abstract

Gas chromatography-mass spectrometry (GC/MS) has long been considered one of the premiere analytical tools for small molecule analysis. Recently, a number of GC/MS systems equipped with high-resolution mass analyzers have been introduced. These systems provide analysts with a new dimension of information – accurate mass measurement to the third or fourth decimal place; however, existing data processing tools do not capitalize on this information. Beyond that, GC/MS spectral reference libraries, which have been curated over the last several decades, contain almost exclusively unit resolution MS spectra making integration of accurate mass data dubious. Here we present an informatic approach, called High-Resolution Filtering (HRF), which bridges this gap. During HRF, high-resolution mass spectra are assigned putative identifications through traditional spectral matching at unit resolution. Once candidate identities have been assigned, all unique combinations of atoms from these candidate precursors are generated and matched to m/z peaks using narrow mass tolerances. The total amount of measured signal that is annotated is used as a metric of plausibility for the presumed identification. Here we demonstrate that the HRF approach is both feasible and highly specific towards correct identifications.

Real time health monitoring through urine metabolomics

“…Metabolite intensity values were normalized to the total ion current (TIC) for the RAW files for each sample using previously described software 21 . Principal component analysis was conducted on log2-transformed TIC-normalized intensity values using the decomposition.PCA() method in Python's skickit learn module 24 .…”

“…Raw files were subsequently processed using in-house software 21 for deconvolution, peak alignment, quantitation, and identification. Cutoffs for peak quantitation were set to a minimum fragment count of 10, minimum observation of a given peak across all files set to 33%, and analyte/background signal set to 10.…”

“…Urine samples were lyopholized, resuspended, and derivatized with solution containing N-Methyl-N-(trimethylsilyl)trifluoroacetamide, and subsequently analyzed with high resolution Gas Chromatography-Fourier Transform Mass Spectrometry (GC-FTMS). The resulting data were deconvoluted with previously described software 21 , which detected and quantified 603 metabolite features across the 109 individual urine samples. Of these 603 metabolite features, 101 metabolites were annotated/identified based on spectral matching (see Methods).…”

“…Samples were dried down, derivatized with n-methyl-n-(trimethylsilyl)trifluoroacetamide, and analyzed with high resolution GC-Fourier Transform Mass Spectrometry (FTMS). (b) Deconvolution and quantification with in-house software21 provide time series profiles for 603 metabolite features. (c) The log2-intensity of a metabolite feature identified as dihydroferulic acid, a known urine metabolite, revealed different baseline concentrations for Subject 1 and Subject 2, compared to a hypothetical range for the general population.…”

Isotopic Ratio Outlier Analysis of the S. cerevisiae Metabolome Using Accurate Mass Gas Chromatography/Time-of-Flight Mass Spectrometry: A New Method for Discovery

“…However, the methods they developed for the GC-Orbitrap MS are not easily adaptable to other accurate mass GC-Time of Flight-MS (GC-TOF/MS) instruments. High Resolution Filtering (HRF) is another GC-Orbitrap based method, without 13 C labeling, that attempts to annotate every measured m/z peak in an EI mass spectrum, leveraging accurate mass to discern between possible identifications 14 .…”