國立高雄大學圖資館 |

登入

回首頁

Computational Methods for Gene Regulation Inference From Single-Cell Multi-Omics Data /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Computational Methods for Gene Regulation Inference From Single-Cell Multi-Omics Data /Elham Jafari.
作者:	Jafari, Elham,
面頁冊數:	1 electronic resource (166 pages)
附註:	Source: Dissertations Abstracts International, Volume: 86-02, Section: B.
附註:	Advisors: Wang, Yijie Committee members: Crandall, David; Tang, Haixu; Zhang, Xuhong.
Contained By:	Dissertations Abstracts International86-02B.
標題:	Computer science.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31332049
ISBN:	9798383574072

Computational Methods for Gene Regulation Inference From Single-Cell Multi-Omics Data /
Jafari, Elham,

Computational Methods for Gene Regulation Inference From Single-Cell Multi-Omics Data /Elham Jafari. - 1 electronic resource (166 pages)

Source: Dissertations Abstracts International, Volume: 86-02, Section: B.

Gene Regulatory Networks (GRNs) map the complex interactions between transcription factors (TFs) and their target genes. GRN inference is essential for understanding cellular dynamics and biological functions. Despite progress, we can further improve current GRN inference methods. Recent advancements in cell profiling technologies enabled simultaneous measurements of gene expression and open chromatin across in thousands of cells. This research offers novel computational pipelines to integrate multi-omics single- cell gene expression and chromatin accessibility data for a multifaceted view of gene regulation as well as enhancing GRN inference through deep learning.Previously, most methods depended solely on gene expression data for GRN inference. This study overcomes such limitations by developing an innovative pipeline for unsupervised integration of single-cell gene expression and chromatin accessibility via their biological consistency. Therefore, we enhance GRN inference by offering a more holistic view of gene regulation. Previous GRN inference solutions rely on various techniques including regression, correlation, information theory, and Ordinary Differential Equations (ODEs). According to system biology belief, gene regulation can be expressed as a system of ODEs, but existing ODE-based methodologies rely on oversimplified linear modeling. To surpass this, we introduce a non-linear neural ODE pipeline that models the dynamic rate of gene expression influenced by TFs in the regulatory network. Lastly, building upon the above holistic multi-omics integration data and non-linear ODE-based GRN inference model, we propose a hybrid GRN refinement strategy. Our approach is based on the biology of regulation involving TFs, genes, open chromatin, and known regulatory regions. We utilize multi-omics data to exclude TFs that are not associated with genes' regulatory regions to ensure biologically relevant regulatory interactions. Overall, my dissertation advances GRN inference research by employing powerful deep learning models over comprehensive multi-omics data, which significantly enhances our understanding of gene regulation.

English

ISBN: 9798383574072Subjects--Topical Terms:

199325
Computer science.
Subjects--Index Terms:

Deep learning

Computational Methods for Gene Regulation Inference From Single-Cell Multi-Omics Data /
LDR:03677nmm a22004453i 4500 001 676343
005 20250420214347.5
006 m o d
007 cr|nu||||||||
008 250901s2024 miu||||||m |||||||eng d
020 $a 9798383574072
035 $a (MiAaPQD)AAI31332049
035 $a AAI31332049
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Jafari, Elham, $e author. $0 (orcid)0000-0001-7511-9266 $3 990126
245 1 0 $a Computational Methods for Gene Regulation Inference From Single-Cell Multi-Omics Data / $c Elham Jafari.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (166 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-02, Section: B.
500 $a Advisors: Wang, Yijie Committee members: Crandall, David; Tang, Haixu; Zhang, Xuhong.
502 $b Ph.D. $c Indiana University $d 2024.
520 $a Gene Regulatory Networks (GRNs) map the complex interactions between transcription factors (TFs) and their target genes. GRN inference is essential for understanding cellular dynamics and biological functions. Despite progress, we can further improve current GRN inference methods. Recent advancements in cell profiling technologies enabled simultaneous measurements of gene expression and open chromatin across in thousands of cells. This research offers novel computational pipelines to integrate multi-omics single- cell gene expression and chromatin accessibility data for a multifaceted view of gene regulation as well as enhancing GRN inference through deep learning.Previously, most methods depended solely on gene expression data for GRN inference. This study overcomes such limitations by developing an innovative pipeline for unsupervised integration of single-cell gene expression and chromatin accessibility via their biological consistency. Therefore, we enhance GRN inference by offering a more holistic view of gene regulation. Previous GRN inference solutions rely on various techniques including regression, correlation, information theory, and Ordinary Differential Equations (ODEs). According to system biology belief, gene regulation can be expressed as a system of ODEs, but existing ODE-based methodologies rely on oversimplified linear modeling. To surpass this, we introduce a non-linear neural ODE pipeline that models the dynamic rate of gene expression influenced by TFs in the regulatory network. Lastly, building upon the above holistic multi-omics integration data and non-linear ODE-based GRN inference model, we propose a hybrid GRN refinement strategy. Our approach is based on the biology of regulation involving TFs, genes, open chromatin, and known regulatory regions. We utilize multi-omics data to exclude TFs that are not associated with genes' regulatory regions to ensure biologically relevant regulatory interactions. Overall, my dissertation advances GRN inference research by employing powerful deep learning models over comprehensive multi-omics data, which significantly enhances our understanding of gene regulation.
546 $a English
590 $a School code: 0093
650 4 $a Computer science. $3 199325
650 4 $a Systematic biology. $3 708586
650 4 $a Bioinformatics. $3 194415
650 4 $a Genetics. $3 189015
653 $a Deep learning
653 $a Gene Regulatory Networks
653 $a Machine learning
653 $a Multi-omics integration
653 $a Single-cell gene expression
690 $a 0984
690 $a 0423
690 $a 0369
690 $a 0715
710 2 $a Indiana University. $b Computer Science. $e degree granting institution. $3 990127
720 1 $a Wang, Yijie $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-02B.
790 $a 0093
791 $a Ph.D.
792 $a 2024
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31332049