Back to search

Purdue University

Design and Statistical Analysis of Mass Spectrometry Imaging Experiments

Abstract

dc:description.abstract

The goal of mass spectrometry-based imaging (MSI) is to characterize the chemical composition of biological samples such as tissues in spatial resolution, and changes in the composition between conditions such as the status of a disease. A single mass spectrometry image is a collection of mass spectra at gridded locations on a tissue. When acquired over multiple biological individuals, tissues, and conditions, the resulting datasets are tremendously complex in both scale and in the stochastic structure of the quantified spectral features. Accurate characterization of the biological and technical variation in these experiments is therefore key for deriving sensitive and reproducible biological conclusions. This dissertation develops a rigorous statistical framework, based on hierarchical Bayesian spatial models, that accurately represents arbitrary complex experimental designs, in (1) structure of the conditions or stresses, (2) the origin of the biological tissues, and (3) the within-tissue spatial dependencies in the quantified spectral features. It makes a methodological contribution by expanding spatial statistics to a large-scale, highly multivariate setting. The performance of the proposed statistical and computational methods are evaluated on a series of synthetic datasets with known stochastic structure, and applied to real-life biomedical experiments. Finally, an open-source R package msiCompare is contributed to provide a practical solution for carrying out the statistical inference in MSI experiments.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy (PhD)
Level thesis:degree_level
Dissertation
Discipline thesis:degree_discipline
Statistics

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Harry, April Janae
Contributors dc:contributor
  • Olga Vitek
  • Mark Ward
  • Hao Zhang

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Purdue University
Base URL
docs.lib.purdue.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Harry, April Janae. Design and Statistical Analysis of Mass Spectrometry Imaging Experiments. Dissertation thesis, https://docs.lib.purdue.edu/open_access_dissertations/1560