東華大學圖書館 |

New Computational and Statistical Approaches to Improve the Analysis of Human Microbiome Data.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	New Computational and Statistical Approaches to Improve the Analysis of Human Microbiome Data./
作者:	Huang, Caizhi David.
面頁冊數:	1 online resource (154 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-01, Section: B.
Contained By:	Dissertations Abstracts International85-01B.
標題:	Mean square errors. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30516301click for full text (PQDT)
ISBN:	9798379870164

New Computational and Statistical Approaches to Improve the Analysis of Human Microbiome Data.
Huang, Caizhi David.

New Computational and Statistical Approaches to Improve the Analysis of Human Microbiome Data. - 1 online resource (154 pages)

Source: Dissertations Abstracts International, Volume: 85-01, Section: B.

Thesis (Ph.D.)--North Carolina State University, 2023.

Includes bibliographical references

The human microbiome is a collection of all microbes that live on and inside us. They closely interact with the human body and have critical impacts on human health. In the past twenty years, human microbiome research has been renewed and evolved with advances in next-generation sequencing technologies (e.g., marker-gene and shotgun metagenomics sequencing), which has deepened our understanding of the human microbiome. Meanwhile, bioinformatics tools and statistical methods have been developed to interpret the microbiome sequencing data and guide microbiome research. The challenge of analyzing microbiome data comes from both the uniqueness of microbiome data features (e.g., compositional, phylogenetically correlated) and the lack of a standard for conducting microbiome research. In this dissertation, we address the following four challenges in microbiome data analysis (corresponding to Chapter 2 to Chapter 5, respectively), with the goal of improving the accuracy and consistency of microbiome research.The measurement of microbial communities using marker-gene and shotgun metagenomic sequencing is affected by contaminant DNA sequences that are not truly present in a sample. In Chapter 2, we propose a hypothesis testing procedure called tcontam to detect contaminant DNA in the marker-gene and metagenomics data. Tcontam is a frequencybased approach that relies on the frequencies of sequence variants and pre-sequencing DNA quantification information for each sample. Compared to the current method, tcontamimproved the performance of detecting contaminants with an easier interpretation of the decision-making and false discovery rate control.In Chapter 3, to address the mixed and inconsistent reported associations between the vaginal microbiome and preterm birth (PTB), we perform a meta-analysis of 12 prospective case-control PTB datasets obtained by using 16S rRNA gene sequencing to measure the vaginal microbiome during pregnancy. Using the machine learning approach, we investigate the predictability of PTB from the composition of the vaginal microbiome in each study and evaluate the cross-dataset reproducibility of PTB predictions from the vaginal microbiome. We further explore the association between specific microbial taxa and PTB within and across studies.The microbiome compositional profile obtained from 16S rRNA gene or metagenomics sequencing is typically high-dimensional, sparse, and compositional, which challenges the traditional association test methods. In Chapter 4, we propose a novel association test called POST to detect the outcome-related operational taxonomic units (OTUs). Compared to existing methods, POST boosts the detecting power of the target OTU by adaptively borrowing information from its phylogenetically close OTUs. Besides, POSTis built on a kernel machine regression model, which can flexibly accommodate complex microbiome effects, be applicable to both continuous and binary outcomes, and easily adjust for covariates.Increasing longitudinal microbiome studies have been conducted to capture the dynamic nature of the microbiome, which causes the need for statistical methods to assess the temporal relationship between specific microbiome taxon and outcome of interest. In Chapter 5, we propose to use the function-to-function regression model to analyze the association between longitudinal microbiome data and longitudinal outcomes. Given the complexity of the longitudinal microbiome data including missing data from irregular time points, the strength of the within-individual correlation, compositionality, and sparsity, we designed simulations to rigorously evaluate the applicability of the function-to-function model for microbiome data and further applied the method to a gut microbiome dataset as a conceptual example.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798379870164Subjects--Topical Terms:

3562318
Mean square errors.
Index Terms--Genre/Form:

542853
Electronic books.

New Computational and Statistical Approaches to Improve the Analysis of Human Microbiome Data.
LDR:05181nmm a2200373K 4500 001 2361121
005 20231024102937.5
006 m o d
007 cr mn ---uuuuu
008 241011s2023 xx obm 000 0 eng d
020 $a 9798379870164
035 $a (MiAaPQ)AAI30516301
035 $a (MiAaPQ)NCState_Univ18402040835
035 $a AAI30516301
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Huang, Caizhi David. $3 3701772
245 1 0 $a New Computational and Statistical Approaches to Improve the Analysis of Human Microbiome Data.
264 0 $c 2023
300 $a 1 online resource (154 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: 85-01, Section: B.
500 $a Advisor: Peng, Xinxia;Kleiner, Manuel;Callahan, Benjamin J.;Tzeng, Jung-Ying.
502 $a Thesis (Ph.D.)--North Carolina State University, 2023.
504 $a Includes bibliographical references
520 $a The human microbiome is a collection of all microbes that live on and inside us. They closely interact with the human body and have critical impacts on human health. In the past twenty years, human microbiome research has been renewed and evolved with advances in next-generation sequencing technologies (e.g., marker-gene and shotgun metagenomics sequencing), which has deepened our understanding of the human microbiome. Meanwhile, bioinformatics tools and statistical methods have been developed to interpret the microbiome sequencing data and guide microbiome research. The challenge of analyzing microbiome data comes from both the uniqueness of microbiome data features (e.g., compositional, phylogenetically correlated) and the lack of a standard for conducting microbiome research. In this dissertation, we address the following four challenges in microbiome data analysis (corresponding to Chapter 2 to Chapter 5, respectively), with the goal of improving the accuracy and consistency of microbiome research.The measurement of microbial communities using marker-gene and shotgun metagenomic sequencing is affected by contaminant DNA sequences that are not truly present in a sample. In Chapter 2, we propose a hypothesis testing procedure called tcontam to detect contaminant DNA in the marker-gene and metagenomics data. Tcontam is a frequencybased approach that relies on the frequencies of sequence variants and pre-sequencing DNA quantification information for each sample. Compared to the current method, tcontamimproved the performance of detecting contaminants with an easier interpretation of the decision-making and false discovery rate control.In Chapter 3, to address the mixed and inconsistent reported associations between the vaginal microbiome and preterm birth (PTB), we perform a meta-analysis of 12 prospective case-control PTB datasets obtained by using 16S rRNA gene sequencing to measure the vaginal microbiome during pregnancy. Using the machine learning approach, we investigate the predictability of PTB from the composition of the vaginal microbiome in each study and evaluate the cross-dataset reproducibility of PTB predictions from the vaginal microbiome. We further explore the association between specific microbial taxa and PTB within and across studies.The microbiome compositional profile obtained from 16S rRNA gene or metagenomics sequencing is typically high-dimensional, sparse, and compositional, which challenges the traditional association test methods. In Chapter 4, we propose a novel association test called POST to detect the outcome-related operational taxonomic units (OTUs). Compared to existing methods, POST boosts the detecting power of the target OTU by adaptively borrowing information from its phylogenetically close OTUs. Besides, POSTis built on a kernel machine regression model, which can flexibly accommodate complex microbiome effects, be applicable to both continuous and binary outcomes, and easily adjust for covariates.Increasing longitudinal microbiome studies have been conducted to capture the dynamic nature of the microbiome, which causes the need for statistical methods to assess the temporal relationship between specific microbiome taxon and outcome of interest. In Chapter 5, we propose to use the function-to-function regression model to analyze the association between longitudinal microbiome data and longitudinal outcomes. Given the complexity of the longitudinal microbiome data including missing data from irregular time points, the strength of the within-individual correlation, compositionality, and sparsity, we designed simulations to rigorously evaluate the applicability of the function-to-function model for microbiome data and further applied the method to a gut microbiome dataset as a conceptual example.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Mean square errors. $3 3562318
650 4 $a Phylogenetics. $3 3561778
650 4 $a Bioinformatics. $3 553671
650 4 $a Premature birth. $3 3681349
650 4 $a Taxonomy. $3 3556303
650 4 $a Genetics. $3 530508
650 4 $a Obstetrics. $3 634501
650 4 $a Systematic biology. $3 3173492
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0715
690 $a 0800
690 $a 0369
690 $a 0380
690 $a 0423
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a North Carolina State University. $3 1018772
773 0 $t Dissertations Abstracts International $g 85-01B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30516301 $z click for full text (PQDT)