UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Peptide detectability algorithm and mass tag labeling, fractionation, and ion acquisition methods in mass spectrometry-based proteomics Riley, Ryan

Abstract

Proteins are responsible for facilitating and regulating nearly all cellular processes, and the collection of all proteins in a biological system is referred to as the proteome. The field of proteomics then aims to both comprehensively and quantitatively understand the dynamics of the proteome, thus providing molecular insights to the functions of biology. This has largely been enabled by mass spectrometry (MS), a technique used to measure gas phase ions of peptides yielded from protein digestion. When designing experiments in proteomics, compromises are typically made between quantitative accuracy and proteome coverage, making it challenging to select experimental strategies and parameters given a diversity of both goals and options. For example, certain targeted strategies rely on synthetic peptides either for use as quantitative standards or for boosting the signal of corresponding endogenous peptides. However, most proteins contain a plethora of theoretical peptide sequences, many of which are not detectable by MS due to non-conducive physiochemical properties. Consequently, the selection of peptides to represent proteins targeted by these strategies is not straightforward. Experimental options are also plentiful when utilizing tandem mass tags (TMT), a labeling strategy that enables multiple samples to be combined into a single MS run. While this is an appealing technique for increasing sample throughput in global experiments, use of TMT introduces significant challenges to data analysis and the impact of mitigation strategies is not trivial to predict. Mitigation strategies include reduction of sample complexity through liquid chromatography and variations to mass spectrometry acquisition methods that dictate how the instrument gathers data from the ions. In chapter 2, we present a random forest model that predicts peptide detectability in mass spectrometry for application to synthetic peptide selection. This includes an R package containing tools for convenient prioritization of peptides for desired proteins and re-training of the model. In chapter 3, we evaluate the impact of alternate sample preparation strategies and instrument acquisition methods on TMT-based global proteomics. Such options include the depth of liquid chromatography, the type of TMT tag, and the MS acquisition methods used.

Item Media

Item Citations and Data

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International