Fourier Transform Based Analysis of Mass Spectra: Disentangling Mass Heterogeneity and Polydispersity
Loading...
Date
2024-12-19
Authors
Swansiger, Andrew
Journal Title
Journal ISSN
Volume Title
Publisher
University of Oregon
Abstract
Understanding the interactions of small molecules with biomolecules and their complexes is fundamental to the clinical interpretation of biological functions and pharmaceutical development. Conversely, these delicate interactions present a multiplexed problem requiring highly specific and sensitive analytical techniques to capture their subtle variances. Advances in soft ionization mass spectrometry (MS) methods such as electrospray ionization (ESI) and desorption electrospray ionization (DESI) have brought together solution phase separation techniques and sensitive gas phase analysis, reducing both sample concentration and purification requirements and enabling fast multiplexed analysis of data-rich biological samples. As the limitations on analyte size and complexity continue to be pushed back by instrumental and experimental innovations, MS deconvolution tools need to continually advance to keep pace with the increased mass heterogeneity and polydispersity of what we can successfully spray. Among current MS deconvolution algorithms, Fourier transform and Gábor transform (FT/GT) provide a consistent and invertible transform for quick recognition of several classes of periodic signal from polydisperse samples, requiring very few a priori assumptions about the sample while extracting the charge and mass information required by other algorithms for accurate modeling of congested mass spectra. The Prell group’s iFAMS software represents the state-of-the-art in Fourier deconvolution of mass spectra, enabling flexible selection of analyte signals from a spectrogram of m/z and frequency to filter out interferent ions. However, assignment of aperiodic mass shifts in data-rich spectra still proves challenging, as they do not produce unique frequency signals, requiring an understanding of previously unutilized aspects of FT/GT deconvolution for mass spectrometry. Additionally, although iFAMS results are highly reproducible, applications of iFAMS data analysis have remained mostly exploratory, as GT lacks a sufficiently high-throughput implementation for analysis of large data sets.
In the first half of this dissertation, a new tool for mass spectrometry Fourier analysis is developed, utilizing the phase angle information from FT/GT for the characterization of small mass variants embedded in polydisperse mediums such as polymers and lipid membranes. The new method of FT/GT macromolecular mass defect (MMD) analysis achieved similar mass accuracy to mass-domain deconvolution methods and is robust to high instrument noise and low mass contaminants, enabling cross-validation of mass-domain deconvolution models. In a workflow complemented with liquid chromatography mass spectrometry, FT/GT MMD analysis enables characterization of polymer reaction intermediates. The second half of the dissertation extends the reproducibility of FT/GT analysis to protein quantitation of MS imaging data from biological tissue, developing a new workflow for batch deconvolution to process tens of thousands of spectra in a few hours. The distinct protein ion patterns generated by GT simplify characterization of brain tissue eluents, while expanding the range of isolatable proteoform signal available for imaging.
This dissertation includes previously published and unpublished co-authored material.
Description
Keywords
Deconvolution, Electrospray Ionization, Fourier Transform, Mass Spectrometry, Polymer Science, Protein Therapeutics