Calculation of retention time tolerance windows with absolute confidence from shared liquid chromatographic retention data

Abstract

Compound identification by liquid chromatography-mass spectrometry (LC-MS) is a tedious process, mainly because authentic standards must be run on a user’s system to be able to confidently reject a potential identity from its retention time and mass spectral properties. Instead, it would be preferable to use shared retention time/index data to narrow down the identity, but shared data cannot be used to reject candidates with an absolute level of confidence because the data are strongly affected by differences between HPLC systems and experimental conditions. However, a technique called “retention projection” was recently shown to account for many of the differences. In this manuscript, we discuss an approach to calculate appropriate retention time tolerance windows for projected retention times, potentially making it possible to exclude candidates with an absolute level of confidence, without needing to have authentic standards of each candidate on hand. In a range of multi-segment gradients and flow rates run among seven different labs, the new approach calculated tolerance windows that were significantly more appropriate for each retention projection than global tolerance windows calculated for retention projections or linear retention indices. Though there were still some small differences between the labs that evidently were not taken into account, the calculated tolerance windows only needed to be relaxed by 50% to make them appropriate for all labs. Even then, 42% of the tolerance windows calculated in this study without standards were narrower than those required by WADA for positive identification, where standards must be run contemporaneously.

Retention projection enables accurate calculation of liquid chromatographic retention times across labs and methods

“…In the future, it will be important to either correct retention projections for this change in selectivity and/or to have column suitability checks built into the workflow to determine if a particular column is suitable for projections of charged compounds. An example of such a suitability check is described in the following, companion manuscript [41]. We also expect the accuracy of retention projections for all compounds (charged and uncharged) could be improved by taking into account distortion of the gradient from the column (“solvent demixing”) [21,22], pressure effects [23], and frictional (viscous) heating of the column [24].…”

Software Tools and Approaches for Compound Identification of LC-MS/MS Data in Metabolomics

“…By massively increasing the structural scaffold space, the retention model can become more robust, even if these molecules will never be annotated in biological samples. Many retention time prediction models are usually locked to a specific LC column and a solvent and buffer system, unless a “retention projection” method can be applied to transfer data to other chromatographic systems [ 112 , 113 , 114 ].…”

Mass Spectrometry Fingerprints of Small-Molecule Metabolites in Biofluids: Building a Spectral Library of Recurrent Spectra for Urine Analysis

“…For example, the extracted peaks from the LC–MS analysis of seven samples of different urine SRMs using the Ultimate 3000/Orbitrap Fusion-Lumos LC–MS system, only 4327 out of 63818 extracted ions from all samples have RSD less than 20%. These significant intensity fluctuations and the fact that the reproducibility of retention times across different chromatographic setups and methods is still poor − make clear the need for implementing data processing and analysis tools for avoiding uncertainty, particularly regarding low abundance components. The ARUS library provides a reliable tool to make comparisons between samples and laboratories.…”