Boston University
Statistical methods in mitochondrial genetics and proteomic network modeling
Abstract
dc:description.abstractAdvances in high-throughput technologies have enabled the generation of molecular data at an unprecedented scale. Recent sequencing technologies have made it feasible to study low-frequency mitochondrial DNA heteroplasmy across many individuals, while large biobank-based proteomic studies now allow the measurement of thousands of circulating proteins in population cohorts. These developments have created new opportunities to analyze genetic and proteomic data in relation to human disease, but the resulting data structures also present distinct statistical challenges. This dissertation presents three methodological studies in statistical genetics and proteomic data analysis. In Project 1, we develop a family-based framework for association testing of mitochondrial DNA heteroplasmy in pedigree studies using a generalized linear mixed model that accounts for both nuclear kinship and maternal lineage structure. In Project 2, we construct genome-based prediction models for circulating proteins in the UK Biobank and decompose measured protein levels into genetically predicted and non-genetic components. In Project 3, we propose HALO, a hub-adaptive Laplacian-regularized regression method for high-dimensional data with network structure, to improve variable selection and interpretability. Collectively, these projects develop statistical tools for analyzing structured and high-dimensional data and illustrate their use in studies of human disease.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Biostatistics
- Grantor
- Boston University
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Liu, Xue
- Advisor dc:contributor.advisor
-
- Liu, Chunyu
Subjects
dc:subject × 7Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/2144/53589