Leveraging bioinformatics methods to examine sex-related variability in genetic and transcriptomic data
Abstract/Contents
- Abstract
- Women are at more than 1.5-fold higher risk for clinically-relevant adverse drug events. While this higher prevalence is partially due to women being overdosed, research suggests that biological sex differences play a role. However, the biological mechanisms behind how sex differences impact drug response are poorly understood. Gene expression and genetic data have the potential to help us investigate these effects, but many studies and samples neglect to include sex labels or examine the impact of sex. Simultaneously, another challenge stems from the overemphasis of certain sex- specific associations without sufficient replication or consideration of other covariates. Consequently, strategic use of data analytic methods is essential for studying sex-related effects. In this thesis, I argue that novel bioinformatics approaches have the potential to improve our understanding of sex as a variable in genetic and gene expression data, and I describe three con- tributions toward this effort. First, I developed a Bayesian Mixture Model approach for identifying sex-specific genetic effects in Genome Wide Association Studies (GWAS). By applying this model to UK Biobank biomarker data, I found that the majority of these clinical lab values do not show sex differences in their genetics, with the exception of testosterone. Second, I addressed the problem of missing sex labels in gene expression data by training models to infer sample sex, and applying these models to over 600,000 publicly available human and mouse samples. I then used these labels to assess sex bias at scale and in drug studies, examine the complexity of cell line sex, and identify mislabeled samples. Finally, I applied our sex labeling models to smoking-related gene expression data, compared the contributions of sex and smoking to expression, and highlighted limitations in our ability to examine sex-differential smoking interaction effects.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2021; ©2021 |
| Publication date | 2021; 2021 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Flynn, Emily Rebekah |
|---|---|
| Degree supervisor | Altman, Russ |
| Thesis advisor | Altman, Russ |
| Thesis advisor | Khatri, Purvesh |
| Thesis advisor | Rivas, Manuel (Manuel A.) |
| Thesis advisor | Stefanick, Marcia Lynn |
| Degree committee member | Khatri, Purvesh |
| Degree committee member | Rivas, Manuel (Manuel A.) |
| Degree committee member | Stefanick, Marcia Lynn |
| Associated with | Stanford University, Department of Biomedical Informatics |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Emily Flynn. |
|---|---|
| Note | Submitted to the Department of Biomedical Informatics. |
| Thesis | Thesis Ph.D. Stanford University 2021. |
| Location | https://purl.stanford.edu/wb963pr7780 |
Access conditions
- Copyright
- © 2021 by Emily Rebekah Flynn
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...