Multi-tissue differential expression analyses between proxy case and control samples Open Access
Liu, Yutong (Spring 2023)
Abstract
Background: The Polygenic Risk Score (PRS) is a widely adopted approach to amalgamate information from multiple genetic locations to determine the risk of an individual developing a specific medical condition, including Alzheimer's disease (AD). AD is a heritable disease, with over 50 risk loci being identified with genome-wide significance. Computational methods, such as machine learning and artificial intelligence-driven meta-analysis, are being utilized to reveal more novel genes linked to AD, leveraging data from genome-wide association studies (GWAS).
Objectives: This study aims to explore the potential of polygenic risk score (PRS) in defining proxy cases and control samples and identify potential biomarkers based on the proxy case and control samples.
Methods: PRS scores are generated for each individual in Genotype-Tissue Expression (GTEx) database by weighting their genotype at each SNP by the corresponding log odds ratios provided in the PRS. Differential expression analysis is conducted between groups with upper 10% PRS and lower 10% PRS, and genes that exhibit significant up or down-regulation are identified (adjusted p-value < 0.05). The DisGeNET database is utilized to conduct chi-square tests to evaluate the extent of enrichment of AD-related genes that were identified through differential expression analysis in each tissue.
Results: PRS scores calculated for participants in GTEx are normally distributed, with a mean of 2.63 and a standard deviation of 0.33. The differential expression analysis across all tissues identifies 170 genes with significant up or down-regulation, with 154 in brain tissues and 18 in whole blood. Several highly significant genes are found related to AD, including MT-TL2, POMC, HTR2C, CARTPT, KLHL7-AS1, COL24A1, and SOX14. Chi-square tests revealed that AD-related genes are significantly enriched by differential expression analysis stratified by PRS in 12 of 14 tissues.
Conclusions: PRS is useful in defining proxy cases and control samples in AD. Analysis of 13 brain tissues reveals several genes with substantially different expression levels in individuals with higher versus lower PRS scores. PRS is effective in enriching AD-related genes across multiple tissues, while the inclusion of unrelated genes with high significance suggests the need for the further development of PRS.
Table of Contents
Table of Contents
1. Introduction
2. Methods
2.1 Data Source
2.2 PRS calculation
2.3 Differential Expression Analysis
2.4 Database Search and Chi-Square Test
3. Results
3.1 Distribution of PRS Score
3.2 Differential Expression Analysis in Each Tissue
3.3 Database Search and Chi-Square Test
4. Discussion
Reference
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Keyword | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Multi-tissue differential expression analyses between proxy case and control samples () | 2023-04-10 10:24:21 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|