본문

서브메뉴

Statistical Integrative Analysis of Genomic Data
Statistical Integrative Analysis of Genomic Data
Statistical Integrative Analysis of Genomic Data

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20260202104802
ISBN  
9798291554265
DDC  
574
저자명  
Liu, Chuwen.
서명/저자  
Statistical Integrative Analysis of Genomic Data
발행사항  
[Sl] : The University of North Carolina at Chapel Hill, 2025
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2025
형태사항  
142 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-02, Section: B.
주기사항  
Advisor: Wu, Di;Zheng, Xiaojing.
학위논문주기  
Thesis (Ph.D.)--The University of North Carolina at Chapel Hill, 2025.
초록/해제  
요약Understanding the association between phenotypes and multi-omics data has garnered growing interest. When multiple data sources are available, integrative analysis can enhance power and interpretation. In this dissertation, we develop three methods to assess such associations through joint modeling of different types of omics data.In Chapter 2, we propose a novel metric-the biological ratio of metatranscriptomics to metagenomics-to quantify relative activity in microbiome data. We introduce a Gaussian mixture regression model with hypothesis testing, applicable to both cross-sectional and longitudinal data. This framework addresses differential expression under the influence of metagenomic variation and employs a two-step strategy to link microbial taxa to disease. Our method shows strong Type I error control and competitive power in both simulations and real datasets.Chapter 3 introduces a joint Bayesian framework that integrates single-cell and bulk RNA-seq data to estimate cell-type-specific (CTS) differential expression. The model employs a negative binomial likelihood for bulk RNA-seq and a zero-inflated negative binomial for single-cell data. Parameters are estimated via Hamiltonian Monte Carlo, and posterior p-values are used for CTS-DE inference. We demonstrate improved accuracy through simulations and apply the method to paired COVID-19 bulk and single-cell RNA-seq data, identifying novel CTS-differential genes and pathways.Building on Chapter 2, Chapter 4 addresses the challenge of data sparsity in microbiome analysis by proposing a deep probabilistic imputation framework. Designed for paired metagenomic and metatranscriptomic data, the model explicitly handles zero inflation and overdispersion using zero-inflated negative binomial distributions. It incorporates modality-specific encoders and a probabilistic canonical correlation layer to infer shared and private latent spaces. This end-to-end framework not only improves data recovery but also enhances downstream analyses such as differential abundance detection.
일반주제명  
Biostatistics
일반주제명  
Bioinformatics
일반주제명  
Molecular biology
일반주제명  
Genetics
키워드  
Genomic data
키워드  
Metatranscriptomics
키워드  
Cell-type-specific
키워드  
Differential expression
키워드  
Microbiomes
기타저자  
The University of North Carolina at Chapel Hill Biostatistics
기본자료저록  
Dissertations Abstracts International. 87-02B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2025        us                              c    eng  d
■001000017358862
■00520260202104802
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798291554265
■035    ▼a(MiAaPQ)AAI32164693
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a574
■1001  ▼aLiu,  Chuwen.
■24510▼aStatistical  Integrative  Analysis  of  Genomic  Data
■260    ▼a[Sl]▼bThe  University  of  North  Carolina  at  Chapel  Hill▼c2025
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2025
■300    ▼a142  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-02,  Section:  B.
■500    ▼aAdvisor:  Wu,  Di;Zheng,  Xiaojing.
■5021  ▼aThesis  (Ph.D.)--The  University  of  North  Carolina  at  Chapel  Hill,  2025.
■520    ▼aUnderstanding  the  association  between  phenotypes  and  multi-omics  data  has  garnered  growing  interest.  When  multiple  data  sources  are  available,  integrative  analysis  can  enhance  power  and  interpretation.  In  this  dissertation,  we  develop  three  methods  to  assess  such  associations  through  joint  modeling  of  different  types  of  omics  data.In  Chapter  2,  we  propose  a  novel  metric-the  biological  ratio  of  metatranscriptomics  to  metagenomics-to  quantify  relative  activity  in  microbiome  data.  We  introduce  a  Gaussian  mixture  regression  model  with  hypothesis  testing,  applicable  to  both  cross-sectional  and  longitudinal  data.  This  framework  addresses  differential  expression  under  the  influence  of  metagenomic  variation  and  employs  a  two-step  strategy  to  link  microbial  taxa  to  disease.  Our  method  shows  strong  Type  I  error  control  and  competitive  power  in  both  simulations  and  real  datasets.Chapter  3  introduces  a  joint  Bayesian  framework  that  integrates  single-cell  and  bulk  RNA-seq  data  to  estimate  cell-type-specific  (CTS)  differential  expression.  The  model  employs  a  negative  binomial  likelihood  for  bulk  RNA-seq  and  a  zero-inflated  negative  binomial  for  single-cell  data.  Parameters  are  estimated  via  Hamiltonian  Monte  Carlo,  and  posterior  p-values  are  used  for  CTS-DE  inference.  We  demonstrate  improved  accuracy  through  simulations  and  apply  the  method  to  paired  COVID-19  bulk  and  single-cell  RNA-seq  data,  identifying  novel  CTS-differential  genes  and  pathways.Building  on  Chapter  2,  Chapter  4  addresses  the  challenge  of  data  sparsity  in  microbiome  analysis  by  proposing  a  deep  probabilistic  imputation  framework.  Designed  for  paired  metagenomic  and  metatranscriptomic  data,  the  model  explicitly  handles  zero  inflation  and  overdispersion  using  zero-inflated  negative  binomial  distributions.  It  incorporates  modality-specific  encoders  and  a  probabilistic  canonical  correlation  layer  to  infer  shared  and  private  latent  spaces.  This  end-to-end  framework  not  only  improves  data  recovery  but  also  enhances  downstream  analyses  such  as  differential  abundance  detection.
■590    ▼aSchool  code:  0153.
■650  4▼aBiostatistics
■650  4▼aBioinformatics
■650  4▼aMolecular  biology
■650  4▼aGenetics
■653    ▼aGenomic  data
■653    ▼aMetatranscriptomics
■653    ▼aCell-type-specific
■653    ▼aDifferential  expression
■653    ▼aMicrobiomes
■690    ▼a0308
■690    ▼a0369
■690    ▼a0307
■690    ▼a0715
■71020▼aThe  University  of  North  Carolina  at  Chapel  Hill▼bBiostatistics.
■7730  ▼tDissertations  Abstracts  International▼g87-02B.
■790    ▼a0153
■791    ▼aPh.D.
■792    ▼a2025
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17358862▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF18532 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.