Please use this identifier to cite or link to this item:
|Title:||Evaluating the Predictiveness of Polygenic Scores in the Fragile Families and Child Wellbeing Study Using a Multi-Ancestral Genotyping Array|
|Abstract:||Polygenic scores (PGSs), individual-level predictions for acquiring a trait or disease, are poised to advance biomedical outcomes via precision medicine. However, a major challenge surrounding this biomarker is that PGSs have much greater predictive accuracy for European populations than for other ancestries. With improved multi-ancestral genotyping arrays such as Illumina’s Multi-Ethnic Genotyping Array (MEGA), it is necessary to reassess PGS predictiveness across different ancestries. To examine the potential magnitude of performance increase across African and Hispanic cohorts from Illumina’s PsychChip to MEGA, this study constructed and analyzed PGSs using genotypic and phenotypic data from the Fragile Families and Child Wellbeing Study cohort and summary statistics from previous genome-wide association studies (GWASs). I find no discrepancies in the predictive accuracies of PGSs of European children and those of non-European children when using PsychChip genotype data. I also identify a significant difference in PGS performance between Hispanic and African individuals, with the former having more predictive scores. Furthermore, I find that when using MEGA genotype data, predictive accuracies of PGSs are not improved for European, African, or Hispanic cohorts, even when ancestry-relevant SNPs are included in PGS construction. These findings provide a critical step towards understanding the predictive performance of PGSs from multi-ancestral genotyping arrays and GWASs and may have significant implications for the study of PGSs and their consequences in precision medicine and healthcare as a whole.|
|Type of Material:||Princeton University Senior Theses|
|Appears in Collections:||Molecular Biology, 1954-2021|
Files in This Item:
|ZHU-ALEXANDER-THESIS.pdf||4.3 MB||Adobe PDF||Request a copy|
Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.