Visualization and application of amino acid retention coefficients obtained from modeling of peptide retention. Issue 18 (4th September 2018)

• Record Type: Journal Article
• Authors: Mohammed, Yassene; Palmblad, Magnus
• Is Part Of: Journal of separation science. Volume 41:Issue 18(2018)
• Page Start: 3644
• Page End: 3653
• Publication Date: 2018-09-04
• Subjects: mass spectrometry -- proteomics -- retention time modeling -- one‐dimensional separation -- two‐dimensional separation; Separation (Technology) -- Periodicals; Chromatographic analysis -- Periodicals; 543.089
• DOI: 10.1002/jssc.201800488
• Languages: English
• ISSNs: 1615-9306

Abstract

Abstract: We introduce a method for data inspection in liquid separations of peptides using amino acid retention coefficients and their relative change across experiments. Our method allows for the direct comparison between actual experimental conditions, regardless of sample content and without the use of internal standards. The modeling uses linear regression of peptide retention time as a function of amino acid composition. We demonstrate the pH dependency of the model in a control experiment where the pH of the mobile phase was changed in controlled way. We introduce a score to identify the false discovery rate on peptide spectrum match level that corresponds to the set of most robust models, i.e. to maximize the shared agreement between experiments. We demonstrate the method utility in reversed‐phase liquid chromatography using 24 datasets with minimal peptide overlap. We apply our method on datasets obtained from a public repository representing various separation designs, including one‐dimensional reversed‐phase liquid chromatography followed by tandem mass spectrometry, and two‐dimensional online strong cation exchange coupled to reversed‐phase liquid chromatography followed by tandem mass spectrometry, and highlight new insights. Our method provides a simple yet powerful way to inspect data quality, in particular for multidimensional separations, improving comparability of data at no additional experimental cost.