Advancing The Genetic Landscape Of Autoimmune Diseases: A Focus On X-Chromosome Genetics And Preclinical To Clinical Disease Progression
Restricted (Penn State Only)
- Author:
- Markus, Havell
- Graduate Program:
- Bioinformatics and Genomics (PhD)
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- January 09, 2025
- Committee Members:
- David Koslicki, Program Head/Chair
Dajiang Liu, Chair & Dissertation Advisor
Qunhua Li, Outside Unit Member
Laura Carrel, Major Field Member & Dissertation Advisor
Aron Lukacher, Outside Field Member
Nancy J Olsen, Special Member - Keywords:
- X chromosome
GWAS
Polygenic risk score
bioinformatics
computational biology - Abstract:
- Genome wide association studies (GWAS) have identified numerous genetic variants associated with autoimmune diseases, yet most exclude the X chromosome and neglect the progression from pre-clinical disease stage. Given the unique biology of X chromosome inactivation (XCI) and importance of developing biomarkers for disease progression, we hypothesize that: 1) XCI skewing patterns of easily accessible tissues can be utilized to predict skewed XCI of hard-to-biopsy disease-relevant tissues; 2) variable XCI escape is influenced by genetics and contributes to disease risk, particularly in sex-biased disorders; and 3) case-control GWAS can be integrated with limited pre-clinical data to improve polygenic prediction of disease progression. We conducted three studies to test these hypotheses. First, we measured XCI skewing patterns in 5,216 bulk RNA-seq samples in the Genotype-Tissue Expression (GTEx) project from 281 donors and 52 tissue types. We observe significant variability in cross-tissue XCI skewing correlations within an individual. To begin understanding environmental factors that may contribute to the divergence of cross-tissue XCI skewing patterns, we illustrate age, smoking, drinking, and obesity impacts the prevalence of skewed XCI in a tissue-specific manner. Lastly, we demonstrate that utilizing both environmental factors and XCI skewing patterns from multiple easily accessible tissues can significantly improve prediction accuracy of skewed XCI in some hard-to-biopsy tissues but not of others. Second, we develop ESCAPE-FM+, a novel statistical method that infers XCI states in an individual irrespective of XCI mosaicism or the presence of expressed heterozygous SNPs. We applied ESCAPE-FM+ to the GTEx project, constructing a comprehensive atlas of XCI escape genes across human tissues. We identify an expanded set of genes exhibiting tissue-specific-escape and observe significant variability in XCI escape both within and between individuals. We lastly analyzed XCI escape as a trait and identified and replicated 109 genes with significant XCI escape heritability in at least one tissue, quantifying genetic influences on variable XCI escape for the first time. Third, we propose a novel method Genetic Progression Score (GPS), which incorporates PRS weights from case-control studies as prior to improve prediction accuracy. In simulations, GPS consistently yields better prediction accuracy than alternative strategies. We derive PRS for the progression from preclinical rheumatoid arthritis and systemic lupus erythematosus in the BioVU biobank and validate them in All of Us. For both diseases, GPS achieves the highest prediction R^2 and the resulting PRS yields the strongest correlation with progression prevalence. Lastly, in PheWAS analysis we show GPS is more specific to disease progression when compared to alternative methods. Overall, the findings presented in these three studies improve our understanding of autoimmune disease pathogenesis, especially regarding the role of genetics in variable XCI escape and progression from pre-clinical disease.