서브메뉴
검색
Decoding Cancer Alterations Through Machine Learning and Interactive Data Visualization
Decoding Cancer Alterations Through Machine Learning and Interactive Data Visualization
Detailed Information
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211151447
- ISBN
- 9798382807485
- DDC
- 574
- 서명/저자
- Decoding Cancer Alterations Through Machine Learning and Interactive Data Visualization
- 발행사항
- [Sl] : Princeton University, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 83 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
- 주기사항
- Advisor: Singh, Mona.
- 학위논문주기
- Thesis (Ph.D.)--Princeton University, 2024.
- 초록/해제
- 요약The rapid proliferation of large-scale cancer sequencing data and the subsequent development of computational methods to analyze it has afforded us a unique opportunity to interrogate the fundamental processes disrupted in cancer. However, there remain many unresolved questions in cancer biology, including: the complex interplay between changes in the cancer genome and transcriptome with downstream changes in the metabolome, which has shown to play a crucial role in a variety of cancer contexts; and the impact of variants of unknown significance which arise in clinical tumor sequencing panels. Both of these outstanding questions hold a particular importance in the development of targeted therapeutics and precision cancer medicine. In this dissertation, I introduce novel computational approaches to advance our understanding of cancer biology on these fronts. First, I train a machine learning method to predict metabolite levels in cancer from gene expression data. I employ a sophisticated cross-validation procedure to maximize power and minimize bias, apply state-of-the-art feature importance analysis to uncover underlying gene-metabolite relationships behind model predictions, and demonstrate that the methodology is broadly applicable to both cancer cell line and patient-derived primary tumor data. Next, I develop a library of aggregated features to characterize cancer somatic mutations with the aim of identifying driver events. I employ this library in the development of an interactive visualization tool to aid in the interrogation of variants of unknown significance in individual patient sequencing data. I then demonstrate the utility of this tool in distinguishing mutations with known cancer effects in a large cohort of targeted tumor sequencing patients. Together, these two approaches provide deeper insights into cancer biology and will enable the development of more effective therapies to treat cancer in the clinic.
- 일반주제명
- Bioinformatics
- 일반주제명
- Computer science
- 일반주제명
- Biology
- 일반주제명
- Oncology
- 키워드
- Cancer biology
- 키워드
- Cancer genomics
- 키워드
- Machine learning
- 키워드
- Metabolomics
- 키워드
- Transcriptomics
- 기타저자
- Princeton University Quantitative Computational Biology
- 기본자료저록
- Dissertations Abstracts International. 85-12B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017161804
■00520250211151447
■006m o d
■007cr#unu||||||||
■020 ▼a9798382807485
■035 ▼a(MiAaPQ)AAI31296546
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a574
■1001 ▼aMuscarella, Antonio Dominic.▼0(orcid)0000-0001-8399-4378
■24510▼aDecoding Cancer Alterations Through Machine Learning and Interactive Data Visualization
■260 ▼a[Sl]▼bPrinceton University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a83 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-12, Section: B.
■500 ▼aAdvisor: Singh, Mona.
■5021 ▼aThesis (Ph.D.)--Princeton University, 2024.
■520 ▼aThe rapid proliferation of large-scale cancer sequencing data and the subsequent development of computational methods to analyze it has afforded us a unique opportunity to interrogate the fundamental processes disrupted in cancer. However, there remain many unresolved questions in cancer biology, including: the complex interplay between changes in the cancer genome and transcriptome with downstream changes in the metabolome, which has shown to play a crucial role in a variety of cancer contexts; and the impact of variants of unknown significance which arise in clinical tumor sequencing panels. Both of these outstanding questions hold a particular importance in the development of targeted therapeutics and precision cancer medicine. In this dissertation, I introduce novel computational approaches to advance our understanding of cancer biology on these fronts. First, I train a machine learning method to predict metabolite levels in cancer from gene expression data. I employ a sophisticated cross-validation procedure to maximize power and minimize bias, apply state-of-the-art feature importance analysis to uncover underlying gene-metabolite relationships behind model predictions, and demonstrate that the methodology is broadly applicable to both cancer cell line and patient-derived primary tumor data. Next, I develop a library of aggregated features to characterize cancer somatic mutations with the aim of identifying driver events. I employ this library in the development of an interactive visualization tool to aid in the interrogation of variants of unknown significance in individual patient sequencing data. I then demonstrate the utility of this tool in distinguishing mutations with known cancer effects in a large cohort of targeted tumor sequencing patients. Together, these two approaches provide deeper insights into cancer biology and will enable the development of more effective therapies to treat cancer in the clinic.
■590 ▼aSchool code: 0181.
■650 4▼aBioinformatics
■650 4▼aComputer science
■650 4▼aBiology
■650 4▼aOncology
■653 ▼aCancer biology
■653 ▼aCancer genomics
■653 ▼aCancer metabolism
■653 ▼aMachine learning
■653 ▼aMetabolomics
■653 ▼aTranscriptomics
■690 ▼a0715
■690 ▼a0984
■690 ▼a0306
■690 ▼a0992
■71020▼aPrinceton University▼bQuantitative Computational Biology.
■7730 ▼tDissertations Abstracts International▼g85-12B.
■790 ▼a0181
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17161804▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
Preview
Export
ChatGPT Discussion
AI Recommended Related Books
Подробнее информация.
- Бронирование
- не существует
- моя папка
- Первый запрос зрения
- Non-Book Loan Application
- Nighttime Book Loan Application
Available after logging in.


