서브메뉴
검색
Contextual Learning on Graphs for Precision Medicine
Contextual Learning on Graphs for Precision Medicine
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211152827
- ISBN
- 9798346567400
- DDC
- 574
- 저자명
- Li, Michelle M.
- 서명/저자
- Contextual Learning on Graphs for Precision Medicine
- 발행사항
- [Sl] : Harvard University, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 371 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-05, Section: B.
- 주기사항
- Advisor: Zitnik, Marinka.
- 학위논문주기
- Thesis (Ph.D.)--Harvard University, 2024.
- 초록/해제
- 요약Precision medicine requires reasoning over interconnected data across multiple modalities to tailor medical decisions based on the context of individual patients. Graphs---or networks---are universal descriptors for systems of interacting elements, and deep learning on biomedical graphs has facilitated advancements in medicine, including accelerated disease gene prioritization and drug target identification. However, existing graph-based models are context-free: unable to adjust their outputs based on the contexts in which they operate. This PhD dissertation innovates two fundamental contextual learning algorithms, SHEPHERD and PINNACLE, to tackle medical questions for which patient and cell type contexts, respectively, are important. SHEPHERD addresses the challenge of low sample sizes among rare diseases by infusing patient data with external biomedical knowledge. It considers individual patients as unique subgraphs in a rare disease knowledge graph to learn patient-specific contexts derived from relationships such as genotype-phenotype and disease-gene associations, phenotype ontology, and genetic pathways. SHEPHERD's contextualized patient representations are optimized for multi-faceted rare disease diagnosis: performing causal gene discovery, retrieving "patients-like-me" with the same causal gene or disease, and providing interpretable characterizations of novel disease presentations. PINNACLE leverages cell-type-specific gene expression as well as cellular and tissue organization to resolve the role of a protein depending on the cell type context. It generates unique protein representations for every cell type context using cell-type-specific protein interaction networks constructed from single-cell transcriptomic atlases, and enforces the global organization of these representations with a metagraph of cell type communication and tissue hierarchy. PINNACLE's context-aware protein representations enable the analysis of drug effects across cell type contexts and the prediction of therapeutic targets in a cell-type-specific manner. Overall, this PhD dissertation pioneers the development of contextualized models to empower precision medicine, from rare disease diagnosis to drug discovery.
- 일반주제명
- Bioinformatics
- 일반주제명
- Computer science
- 일반주제명
- Health sciences
- 키워드
- Drug discovery
- 기타저자
- Harvard University Medical Sciences
- 기본자료저록
- Dissertations Abstracts International. 86-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017164061
■00520250211152827
■006m o d
■007cr#unu||||||||
■020 ▼a9798346567400
■035 ▼a(MiAaPQ)AAI31560251
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a574
■1001 ▼aLi, Michelle M.▼0(orcid)0000-0003-0223-7485
■24510▼aContextual Learning on Graphs for Precision Medicine
■260 ▼a[Sl]▼bHarvard University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a371 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-05, Section: B.
■500 ▼aAdvisor: Zitnik, Marinka.
■5021 ▼aThesis (Ph.D.)--Harvard University, 2024.
■520 ▼aPrecision medicine requires reasoning over interconnected data across multiple modalities to tailor medical decisions based on the context of individual patients. Graphs---or networks---are universal descriptors for systems of interacting elements, and deep learning on biomedical graphs has facilitated advancements in medicine, including accelerated disease gene prioritization and drug target identification. However, existing graph-based models are context-free: unable to adjust their outputs based on the contexts in which they operate. This PhD dissertation innovates two fundamental contextual learning algorithms, SHEPHERD and PINNACLE, to tackle medical questions for which patient and cell type contexts, respectively, are important. SHEPHERD addresses the challenge of low sample sizes among rare diseases by infusing patient data with external biomedical knowledge. It considers individual patients as unique subgraphs in a rare disease knowledge graph to learn patient-specific contexts derived from relationships such as genotype-phenotype and disease-gene associations, phenotype ontology, and genetic pathways. SHEPHERD's contextualized patient representations are optimized for multi-faceted rare disease diagnosis: performing causal gene discovery, retrieving "patients-like-me" with the same causal gene or disease, and providing interpretable characterizations of novel disease presentations. PINNACLE leverages cell-type-specific gene expression as well as cellular and tissue organization to resolve the role of a protein depending on the cell type context. It generates unique protein representations for every cell type context using cell-type-specific protein interaction networks constructed from single-cell transcriptomic atlases, and enforces the global organization of these representations with a metagraph of cell type communication and tissue hierarchy. PINNACLE's context-aware protein representations enable the analysis of drug effects across cell type contexts and the prediction of therapeutic targets in a cell-type-specific manner. Overall, this PhD dissertation pioneers the development of contextualized models to empower precision medicine, from rare disease diagnosis to drug discovery.
■590 ▼aSchool code: 0084.
■650 4▼aBioinformatics
■650 4▼aComputer science
■650 4▼aHealth sciences
■653 ▼aContextual learning
■653 ▼aDrug discovery
■653 ▼aGraph machine learning
■653 ▼aPrecision medicine
■653 ▼aRare disease diagnosis
■653 ▼aSingle-cell transcriptomics
■690 ▼a0715
■690 ▼a0984
■690 ▼a0566
■71020▼aHarvard University▼bMedical Sciences.
■7730 ▼tDissertations Abstracts International▼g86-05B.
■790 ▼a0084
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164061▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


