東華大學圖書館 |

Novel Statistical Approaches for High-Dimensional Microbiome Data and Integrative Analysis of Single-Arm Clinical Studies and Real-World Data.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Novel Statistical Approaches for High-Dimensional Microbiome Data and Integrative Analysis of Single-Arm Clinical Studies and Real-World Data./
Author:	Sachdeva, Archie.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2022,
Description:	108 p.
Notes:	Source: Dissertations Abstracts International, Volume: 84-03, Section: B.
Contained By:	Dissertations Abstracts International84-03B.
Subject:	Biostatistics. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29320499
ISBN:	9798351427829

Novel Statistical Approaches for High-Dimensional Microbiome Data and Integrative Analysis of Single-Arm Clinical Studies and Real-World Data.
Sachdeva, Archie.

Novel Statistical Approaches for High-Dimensional Microbiome Data and Integrative Analysis of Single-Arm Clinical Studies and Real-World Data. - Ann Arbor : ProQuest Dissertations & Theses, 2022 - 108 p.

Source: Dissertations Abstracts International, Volume: 84-03, Section: B.

Thesis (Ph.D.)--University of Florida, 2022.

This item must not be sold to any third party vendors.

Scientific studies conducted during the last two decades have established the central role of the microbiome in disease and health. High-throughput sequencing technologies have allowed the accurate estimation of microbial composition in specimens collected from human and environmental sites. Microbiome data can be used for the prediction of therapeutic outcomes and detecting microbial taxa associated with two or more sample groups defined by attributes such as disease subtype, geography, or environmental condition. The latter phenomenon is known as differential abundance analysis. The results from these analysis, in turn, help clinical practitioners and researchers diagnose disease and develop new treatments more effectively. However, prediction of therapeutic outcomes and detecting differential abundance is uniquely challenging due to the high dimensionality, collinearity, sparsity, and compositionality of microbiome data. In the first chapter, we develop a parsimonious strategy with optimal prediction performance to identify groups of taxa associated with an outcome of interest. Specifically, we use Bayesian variable selection to obtain a sparse set of taxa predictive of the disease outcome while accommodating the compositional nature of the data. This is achieved by utilizing balances, i.e., scale invariant log-contrasts of the taxa relative abundances. Our approach is applicable to binary and continuous outcomes and can be extended to multi-level outcome variables. We illustrate the approach using HIV data and oral microbiome data, comparing the accuracy of the proposed method with existing techniques. Further, for differential abundance analysis there is a critical need for unified statistical approaches that can directly compare more than two groups and appropriately adjust for covariates. We develop a zero-inflated Bayesian nonparametric (ZIBNP) methodology that meets the multipronged challenges posed by microbiome data and identifies differentially abundant taxa in two or more groups, while also accounting for sample-specific covariates. The proposed hierarchical model flexibly adapts to unique data characteristics, casts the typically high proportion of zeros in a censoring framework, and mitigates high dimensionality and collinearity issues by utilizing the dimension reducing property of the semiparametric Chinese restaurant process. The approach relates the microbiome sampling depths to inferential precision and conforms with the compositional nature of microbiome data. In simulation studies and in the analyses of the CAnine Microbiome during Parasitism (CAMP) dataset on infected and uninfected dogs and the Global Gut microbiome dataset on human subjects belonging to three geographical regions, we compare ZIBNP with established statistical methods for differential abundance analysis in the presence of covariates. In the third chapter, we develop a methodology for integrative analyses of single-arm clinical trials and real world data. In recent years, the idea of augmenting randomized clinical trials data with real world data has emerged as a particularly attractive technique for health organizations and drug developers to accelerate the drug development process. Major regulatory authorities such as the Food and Drug Administration and European Medicines Agency have recognized the potential of utilizing real world data and are advancing toward making regulatory decisions based on real world evidence. In recent years, several statistical methods have been developed for borrowing data from real world sources such as electronic health records, products and disease registries, and claims and billing data. We develop a novel approach to augment single-arm clinical trials with the real world data derived from single or multiple data sources. Furthermore, we illustrate the proposed method in the presence of missing data and conduct simulation studies to evaluate the performance of the proposed method in diverse settings.

ISBN: 9798351427829Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Bayesian inference

Novel Statistical Approaches for High-Dimensional Microbiome Data and Integrative Analysis of Single-Arm Clinical Studies and Real-World Data.
LDR:05309nmm a2200385 4500 001 2394370
005 20240422070838.5
006 m o d
007 cr#unu||||||||
008 251215s2022 ||||||||||||||||| ||eng d
020 $a 9798351427829
035 $a (MiAaPQ)AAI29320499
035 $a AAI29320499
040 $a MiAaPQ $c MiAaPQ
100 1 $a Sachdeva, Archie. $3 3763840
245 1 0 $a Novel Statistical Approaches for High-Dimensional Microbiome Data and Integrative Analysis of Single-Arm Clinical Studies and Real-World Data.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2022
300 $a 108 p.
500 $a Source: Dissertations Abstracts International, Volume: 84-03, Section: B.
500 $a Advisor: Guha, Subharup.
502 $a Thesis (Ph.D.)--University of Florida, 2022.
506 $a This item must not be sold to any third party vendors.
520 $a Scientific studies conducted during the last two decades have established the central role of the microbiome in disease and health. High-throughput sequencing technologies have allowed the accurate estimation of microbial composition in specimens collected from human and environmental sites. Microbiome data can be used for the prediction of therapeutic outcomes and detecting microbial taxa associated with two or more sample groups defined by attributes such as disease subtype, geography, or environmental condition. The latter phenomenon is known as differential abundance analysis. The results from these analysis, in turn, help clinical practitioners and researchers diagnose disease and develop new treatments more effectively. However, prediction of therapeutic outcomes and detecting differential abundance is uniquely challenging due to the high dimensionality, collinearity, sparsity, and compositionality of microbiome data. In the first chapter, we develop a parsimonious strategy with optimal prediction performance to identify groups of taxa associated with an outcome of interest. Specifically, we use Bayesian variable selection to obtain a sparse set of taxa predictive of the disease outcome while accommodating the compositional nature of the data. This is achieved by utilizing balances, i.e., scale invariant log-contrasts of the taxa relative abundances. Our approach is applicable to binary and continuous outcomes and can be extended to multi-level outcome variables. We illustrate the approach using HIV data and oral microbiome data, comparing the accuracy of the proposed method with existing techniques. Further, for differential abundance analysis there is a critical need for unified statistical approaches that can directly compare more than two groups and appropriately adjust for covariates. We develop a zero-inflated Bayesian nonparametric (ZIBNP) methodology that meets the multipronged challenges posed by microbiome data and identifies differentially abundant taxa in two or more groups, while also accounting for sample-specific covariates. The proposed hierarchical model flexibly adapts to unique data characteristics, casts the typically high proportion of zeros in a censoring framework, and mitigates high dimensionality and collinearity issues by utilizing the dimension reducing property of the semiparametric Chinese restaurant process. The approach relates the microbiome sampling depths to inferential precision and conforms with the compositional nature of microbiome data. In simulation studies and in the analyses of the CAnine Microbiome during Parasitism (CAMP) dataset on infected and uninfected dogs and the Global Gut microbiome dataset on human subjects belonging to three geographical regions, we compare ZIBNP with established statistical methods for differential abundance analysis in the presence of covariates. In the third chapter, we develop a methodology for integrative analyses of single-arm clinical trials and real world data. In recent years, the idea of augmenting randomized clinical trials data with real world data has emerged as a particularly attractive technique for health organizations and drug developers to accelerate the drug development process. Major regulatory authorities such as the Food and Drug Administration and European Medicines Agency have recognized the potential of utilizing real world data and are advancing toward making regulatory decisions based on real world evidence. In recent years, several statistical methods have been developed for borrowing data from real world sources such as electronic health records, products and disease registries, and claims and billing data. We develop a novel approach to augment single-arm clinical trials with the real world data derived from single or multiple data sources. Furthermore, we illustrate the proposed method in the presence of missing data and conduct simulation studies to evaluate the performance of the proposed method in diverse settings.
590 $a School code: 0070.
650 4 $a Biostatistics. $3 1002712
653 $a Bayesian inference
653 $a Bayesian nonparametrics
653 $a Clinical trials
653 $a High-dimensional data
653 $a Microbiome data
653 $a Real-world evidence
690 $a 0308
710 2 $a University of Florida. $b Biostatistics. $3 3763841
773 0 $t Dissertations Abstracts International $g 84-03B.
790 $a 0070
791 $a Ph.D.
792 $a 2022
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29320499