서브메뉴
검색
Computational Methods in Functional Prioritization of Polygenic Risk Score Models
Computational Methods in Functional Prioritization of Polygenic Risk Score Models
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211153018
- ISBN
- 9798384046172
- DDC
- 574
- 저자명
- Crone, Bradley.
- 서명/저자
- Computational Methods in Functional Prioritization of Polygenic Risk Score Models
- 발행사항
- [Sl] : University of Michigan, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 122 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-04, Section: B.
- 주기사항
- Advisor: Boyle, Alan.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2024.
- 초록/해제
- 요약Polygenic risk scores (PRS) have emerged as a powerful tool in genetic research, providing a means to quantify an individual's genetic predisposition to complex traits and diseases. Accurate prediction models are reliant on large ancestry-specific genome-wide association study (GWAS) summary statistics, which are lacking for non-European ancestry populations. Transferability of European-derived PRS across ancestries are confounded by differences in linkage disequilibrium and genetic architectures, limiting the predictive power of risk models in non-European ancestries.To address this challenge, my dissertation focuses on leveraging functional genomic information to prioritize European GWAS single nucleotide polymorphisms (SNPs) with potential regulatory effects in calculating PRS for non-European target samples. Functional regulatory mutations are likely shared between ancestries, and selecting GWAS SNPs with strong functional evidence over association signals can improve portability of scoring models. In Chapter 2, I introduce tissue-specific functional prioritization of European GWAS SNPs in trans-ancestral PRS models by leveraging RegulomeDB-derived regulatory annotation models. I show that selecting SNPs prioritized by tissue-specific functional probabilities confers greater accuracy in cross-population predictions than selection of GWAS SNPs by association signal alone. The study introduces a method to isolate and identify common functional regulatory mutations across different ancestral backgrounds, aiding in the refinement of disease risk models targeting under-represented populations.In Chapter 3, I expand on the single tissue prioritization model to encompass functional regulatory mutations from all tissue types significantly enriched for trait heritability. I introduce an algorithm that iteratively constructs and optimizes European-derived functionally-informed PRS models. Cross-population validation in African target samples shows enhanced predictive accuracy and transferability of multi-tissue PRS models compared to single-tissue functional prioritization. The results highlight the importance of incorporating functional influences of all implicated tissues to improve the generalizability of complex trait PRS models across diverse ancestries.In Chapter 4, I provide detailed descriptions of the computational pipelines I developed for the single tissue prioritization approach and the multiple tissue optimization algorithm for trans-ancestral functional PRS modeling. I implemented both established and custom software to construct PRS targeting functionally-prioritized GWAS SNPs in my investigations. These tools were developed with the goal of scientific reproducibility and accessibility, and have been made publicly available to enable future researchers in conducting additional studies on the regulatory impacts of polygenic disease risk.Overall, the methods and models presented in my dissertation have the potential to advance trans-ancestral PRS modeling and enhance the precision of disease risk predictions for underrepresented ancestries.
- 일반주제명
- Bioinformatics
- 일반주제명
- Genetics
- 일반주제명
- Biochemistry
- 일반주제명
- Bioengineering
- 기타저자
- University of Michigan Bioinformatics
- 기본자료저록
- Dissertations Abstracts International. 86-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017164576
■00520250211153018
■006m o d
■007cr#unu||||||||
■020 ▼a9798384046172
■035 ▼a(MiAaPQ)AAI31631562
■035 ▼a(MiAaPQ)umichrackham005624
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a574
■1001 ▼aCrone, Bradley.
■24510▼aComputational Methods in Functional Prioritization of Polygenic Risk Score Models
■260 ▼a[Sl]▼bUniversity of Michigan▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a122 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-04, Section: B.
■500 ▼aAdvisor: Boyle, Alan.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2024.
■520 ▼aPolygenic risk scores (PRS) have emerged as a powerful tool in genetic research, providing a means to quantify an individual's genetic predisposition to complex traits and diseases. Accurate prediction models are reliant on large ancestry-specific genome-wide association study (GWAS) summary statistics, which are lacking for non-European ancestry populations. Transferability of European-derived PRS across ancestries are confounded by differences in linkage disequilibrium and genetic architectures, limiting the predictive power of risk models in non-European ancestries.To address this challenge, my dissertation focuses on leveraging functional genomic information to prioritize European GWAS single nucleotide polymorphisms (SNPs) with potential regulatory effects in calculating PRS for non-European target samples. Functional regulatory mutations are likely shared between ancestries, and selecting GWAS SNPs with strong functional evidence over association signals can improve portability of scoring models. In Chapter 2, I introduce tissue-specific functional prioritization of European GWAS SNPs in trans-ancestral PRS models by leveraging RegulomeDB-derived regulatory annotation models. I show that selecting SNPs prioritized by tissue-specific functional probabilities confers greater accuracy in cross-population predictions than selection of GWAS SNPs by association signal alone. The study introduces a method to isolate and identify common functional regulatory mutations across different ancestral backgrounds, aiding in the refinement of disease risk models targeting under-represented populations.In Chapter 3, I expand on the single tissue prioritization model to encompass functional regulatory mutations from all tissue types significantly enriched for trait heritability. I introduce an algorithm that iteratively constructs and optimizes European-derived functionally-informed PRS models. Cross-population validation in African target samples shows enhanced predictive accuracy and transferability of multi-tissue PRS models compared to single-tissue functional prioritization. The results highlight the importance of incorporating functional influences of all implicated tissues to improve the generalizability of complex trait PRS models across diverse ancestries.In Chapter 4, I provide detailed descriptions of the computational pipelines I developed for the single tissue prioritization approach and the multiple tissue optimization algorithm for trans-ancestral functional PRS modeling. I implemented both established and custom software to construct PRS targeting functionally-prioritized GWAS SNPs in my investigations. These tools were developed with the goal of scientific reproducibility and accessibility, and have been made publicly available to enable future researchers in conducting additional studies on the regulatory impacts of polygenic disease risk.Overall, the methods and models presented in my dissertation have the potential to advance trans-ancestral PRS modeling and enhance the precision of disease risk predictions for underrepresented ancestries.
■590 ▼aSchool code: 0127.
■650 4▼aBioinformatics
■650 4▼aGenetics
■650 4▼aBiochemistry
■650 4▼aBioengineering
■653 ▼aFunctional prioritization
■653 ▼aPolygenic risk scores
■653 ▼aSingle nucleotide polymorphisms
■653 ▼aComputational pipelines
■653 ▼aGenome-wide association study
■690 ▼a0715
■690 ▼a0369
■690 ▼a0202
■690 ▼a0487
■71020▼aUniversity of Michigan▼bBioinformatics.
■7730 ▼tDissertations Abstracts International▼g86-04B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164576▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
