東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Advanced Machine Learning and Artifi...

Chen, Jiawen,

Linked to FindBook

Google Book

Amazon

博客來

Advanced Machine Learning and Artificial Intelligence Methods for Spatial Omics /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Advanced Machine Learning and Artificial Intelligence Methods for Spatial Omics // Jiawen Chen.
Author:	Chen, Jiawen,
Description:	1 electronic resource (160 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-11, Section: B.
Contained By:	Dissertations Abstracts International86-11B.
Subject:	Biostatistics. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31938728
ISBN:	9798315729082

Advanced Machine Learning and Artificial Intelligence Methods for Spatial Omics /
Chen, Jiawen,

Advanced Machine Learning and Artificial Intelligence Methods for Spatial Omics /Jiawen Chen. - 1 electronic resource (160 pages)

Source: Dissertations Abstracts International, Volume: 86-11, Section: B.

Spatial transcriptomics (ST) technology has revolutionized our understanding of biological systems by enabling the simultaneous measurement of gene expression and the preservation of spatial localization within tissues. Given this enriched data source, the central question in spatial omics analysis becomes: How can we efficiently utilize all this information and what important biological question can we answer? In this dissertation, we aim to develop ML and AI methods to address the pressing biological problems in spatial omics data analysis.In the first section, we present POLARIS, a versatile ST analysis method that can perform cell type deconvolution, identify anatomical or functional layer-wise differentially expressed (LDE) genes, and enable cell composition inference from histology images. POLARIS employs transfer learning to extract meaningful features from images of each spot and its surroundings. These features are then input into a deep neural network guided by Bayesian posterior probabilities to estimate cell compositions. This integration significantly enhances accuracy, and allows for predicting cell-type proportions in unmeasured regions and applying to new histology images without corresponding gene expression data.In the second section, we propose StarTrail, a novel ML method that leverage spatial gradients for spatial omics data. StarTrail investigates where and how does the omics feature change through spatial gradients, which map the rate of change of omics features across spatial positions. This approach allows us to identify zones of rapid change, characterized by high spatial gradients, which often correspond to critical biological junctures such as interfaces between lesional/diseased and healthy/normal tissues or regions with distinct functions. StarTrail, filling important gaps in current literature, enables deeper insights into tissue spatial architecture.In the third section, we introduce the Nearest Neighbor Derivative Process (NNDP), a scalable Gaussian Process framework that jointly models spatial processes and their derivatives, resulting a time complexity reduction from O(n3) to O(n). NNDP improves StarTrail in the second section by providing less hyper-parameter tuning, more theoretical support. NNDP robustly and accurately estimates spatial derivatives in various simulated spatial patterns and real data analysis.

English

ISBN: 9798315729082Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Spatial transcriptomics

Advanced Machine Learning and Artificial Intelligence Methods for Spatial Omics /
LDR:03908nmm a22004573i 4500 001 2434566
005 20250912102043.5
006 m o d
007 cr|nu||||||||
008 261011s2025 miu||||||m |||||||eng d
020 $a 9798315729082
035 $a (MiAaPQD)AAI31938728
035 $a AAI31938728
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Chen, Jiawen, $e author. $3 3824581
245 1 0 $a Advanced Machine Learning and Artificial Intelligence Methods for Spatial Omics / $c Jiawen Chen.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2025
300 $a 1 electronic resource (160 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: 86-11, Section: B.
500 $a Advisors: Li, Yun; Li, Didong Committee members: Gupta, Gaorav P.; Li, Xihao; Zhang, Daiwei.
502 $b Ph.D. $c The University of North Carolina at Chapel Hill $d 2025.
520 $a Spatial transcriptomics (ST) technology has revolutionized our understanding of biological systems by enabling the simultaneous measurement of gene expression and the preservation of spatial localization within tissues. Given this enriched data source, the central question in spatial omics analysis becomes: How can we efficiently utilize all this information and what important biological question can we answer? In this dissertation, we aim to develop ML and AI methods to address the pressing biological problems in spatial omics data analysis.In the first section, we present POLARIS, a versatile ST analysis method that can perform cell type deconvolution, identify anatomical or functional layer-wise differentially expressed (LDE) genes, and enable cell composition inference from histology images. POLARIS employs transfer learning to extract meaningful features from images of each spot and its surroundings. These features are then input into a deep neural network guided by Bayesian posterior probabilities to estimate cell compositions. This integration significantly enhances accuracy, and allows for predicting cell-type proportions in unmeasured regions and applying to new histology images without corresponding gene expression data.In the second section, we propose StarTrail, a novel ML method that leverage spatial gradients for spatial omics data. StarTrail investigates where and how does the omics feature change through spatial gradients, which map the rate of change of omics features across spatial positions. This approach allows us to identify zones of rapid change, characterized by high spatial gradients, which often correspond to critical biological junctures such as interfaces between lesional/diseased and healthy/normal tissues or regions with distinct functions. StarTrail, filling important gaps in current literature, enables deeper insights into tissue spatial architecture.In the third section, we introduce the Nearest Neighbor Derivative Process (NNDP), a scalable Gaussian Process framework that jointly models spatial processes and their derivatives, resulting a time complexity reduction from O(n3) to O(n). NNDP improves StarTrail in the second section by providing less hyper-parameter tuning, more theoretical support. NNDP robustly and accurately estimates spatial derivatives in various simulated spatial patterns and real data analysis.
546 $a English
590 $a School code: 0153
650 4 $a Biostatistics. $3 1002712
650 4 $a Cellular biology. $3 3172791
650 4 $a Bioinformatics. $3 553671
650 4 $a Genetics. $3 530508
653 $a Spatial transcriptomics
653 $a Gene expression
653 $a Cell type
653 $a Histology images
653 $a StarTrail
690 $a 0308
690 $a 0379
690 $a 0369
690 $a 0715
710 2 $a The University of North Carolina at Chapel Hill. $b Biostatistics. $e degree granting institution. $3 3817171
720 1 $a Li, Yun $e degree supervisor.
720 1 $a Li, Didong $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-11B.
790 $a 0153
791 $a Ph.D.
792 $a 2025
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31938728