서브메뉴
검색
Engineering Cells Using Artificial Intelligence
Engineering Cells Using Artificial Intelligence
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211152745
- ISBN
- 9798342107389
- DDC
- 570
- 저자명
- Roohani, Yusuf.
- 서명/저자
- Engineering Cells Using Artificial Intelligence
- 발행사항
- [Sl] : Stanford University, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 206 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-04, Section: B.
- 주기사항
- Advisor: Leskovec, Jure;Quake, Stephen.
- 학위논문주기
- Thesis (Ph.D.)--Stanford University, 2024.
- 초록/해제
- 요약Driven by advancements in genome engineering and single-cell profiling, the precise engineering of cells has the potential to transform cancer therapy and regenerative medicine. It overcomes many of the challenges of conventional small-molecule based discovery through increased precision and target-specificity coupled with low toxicity and off-target effects. However, the full promise of this technology remains unrealized due to the vast space of possible cell engineering routes that must be tested in the lab. Artificial intelligence can help address this challenge by simulating cell behavior and conducting virtual genetic perturbation experiments. My work develops three capabilities necessary for AI-driven cell engineering: universal cell representation, perturbation effect prediction and closed-loop experiment design.Using a large foundation model trained on gene expression data from millions of cells, we create a universal cell embedding that can represent any cell from any tissue, species or dataset. We simulate cell behavior by predicting cellular response to multi-gene perturbation, and use gene-gene networks to expand predictions to new untested perturbations. Finally, we develop an AI agent, informed by the scientific literature, to iteratively design lab experiments that guide cells towards a desired phenotype. In summary, my work seeks to develop virtual cells where traditional hypothesis-driven lab experiments are augmented by AI-enabled in-silico experimentation.
- 일반주제명
- CRISPR
- 일반주제명
- Gene expression
- 일반주제명
- Biological products
- 일반주제명
- Design
- 일반주제명
- Genomes
- 일반주제명
- Biology
- 일반주제명
- Genomics
- 일반주제명
- Stem cells
- 일반주제명
- Bioengineering
- 일반주제명
- Bioinformatics
- 일반주제명
- Cellular biology
- 일반주제명
- Genetics
- 기타저자
- Stanford University.
- 기본자료저록
- Dissertations Abstracts International. 86-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017163726
■00520250211152745
■006m o d
■007cr#unu||||||||
■020 ▼a9798342107389
■035 ▼a(MiAaPQ)AAI31520285
■035 ▼a(MiAaPQ)Stanfordjw766pz3938
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a570
■1001 ▼aRoohani, Yusuf.
■24510▼aEngineering Cells Using Artificial Intelligence
■260 ▼a[Sl]▼bStanford University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a206 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-04, Section: B.
■500 ▼aAdvisor: Leskovec, Jure;Quake, Stephen.
■5021 ▼aThesis (Ph.D.)--Stanford University, 2024.
■520 ▼aDriven by advancements in genome engineering and single-cell profiling, the precise engineering of cells has the potential to transform cancer therapy and regenerative medicine. It overcomes many of the challenges of conventional small-molecule based discovery through increased precision and target-specificity coupled with low toxicity and off-target effects. However, the full promise of this technology remains unrealized due to the vast space of possible cell engineering routes that must be tested in the lab. Artificial intelligence can help address this challenge by simulating cell behavior and conducting virtual genetic perturbation experiments. My work develops three capabilities necessary for AI-driven cell engineering: universal cell representation, perturbation effect prediction and closed-loop experiment design.Using a large foundation model trained on gene expression data from millions of cells, we create a universal cell embedding that can represent any cell from any tissue, species or dataset. We simulate cell behavior by predicting cellular response to multi-gene perturbation, and use gene-gene networks to expand predictions to new untested perturbations. Finally, we develop an AI agent, informed by the scientific literature, to iteratively design lab experiments that guide cells towards a desired phenotype. In summary, my work seeks to develop virtual cells where traditional hypothesis-driven lab experiments are augmented by AI-enabled in-silico experimentation.
■590 ▼aSchool code: 0212.
■650 4▼aCRISPR
■650 4▼aGene expression
■650 4▼aBiological products
■650 4▼aDesign
■650 4▼aGenomes
■650 4▼aBiology
■650 4▼aGenomics
■650 4▼aStem cells
■650 4▼aBioengineering
■650 4▼aBioinformatics
■650 4▼aCellular biology
■650 4▼aGenetics
■690 ▼a0389
■690 ▼a0800
■690 ▼a0306
■690 ▼a0202
■690 ▼a0715
■690 ▼a0379
■690 ▼a0369
■71020▼aStanford University.
■7730 ▼tDissertations Abstracts International▼g86-04B.
■790 ▼a0212
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17163726▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


