Activity Number:
|
356
- Innovative Analysis Methods for Various Types of High-Throughput and Heterogeneous Data
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
Abstract #330723
|
Presentation
|
Title:
|
Analysis of Time-Course Microbiota Data Through Longitudinal Linear Combination Test
|
Author(s):
|
Elham Khodayari Moez* and Morteza Hajihosseini and Anita Kozyrskyj and Irina Dinu
|
Companies:
|
University of Alberta and University of Alberta and University of Alberta and University of Alberta
|
Keywords:
|
Microbiota;
longitudinal;
high-dimensional data;
gut microbiome;
infancy
|
Abstract:
|
Objective: Incorporating longitudinal analysis designs in studies of the infant gut microbiome advances our understanding of its development during infancy. Although there has been progress in developing analytical methods for these high dimensional time-course datasets, most suffer from serious limitations. Method: A two-step method will be developed to test differential patterns of time-course high-dimensional infant gut microbial abundance data in association with multiple birth and postnatal exposure variables. Either polynomial models or B-splines will be used to accommodate within-subject variation in the first step and Linear Combination Test (LCT) utilized to analyze between-subject variation in the second step. The performance of the method will be evaluated in both simulation and application studies of microbiome data. Results: The simulation study supports the method's efficiency in controlling errors and handling high-dimensionality of the data and small number of infant fecal sampling times. Conclusion: The outstanding strength of the proposed method is its wide applicability to the analysis of multiple continuous and binary predictors.
|
Authors who are presenting talks have a * after their name.