A Case Study Tested Framework for Multivariate Analyses of Microbiomes: Software for Microbial Community Comparisons
The study of microbiomes is important because our understanding of microbial communities is providing insight into human health and many other areas of interest. Researchers often use genomic data to study microbial organisms, demonstrating differences from one organism to the next. Metagenomic data...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | unknown |
| Published: |
University of Montana
2015
|
| Subjects: | |
| Online Access: | https://scholarworks.umt.edu/etd/4554 https://scholarworks.umt.edu/context/etd/article/5572/viewcontent/20151001_Spaulding_Thesis_Finished_Revisions.pdf |
| Summary: | The study of microbiomes is important because our understanding of microbial communities is providing insight into human health and many other areas of interest. Researchers often use genomic data to study microbial organisms, demonstrating differences from one organism to the next. Metagenomic data is utilized to study communities of microbial organisms. The research described herein involved the development of a collection of computational methods. This suite of computational methods and tools (written in the R and Perl languages) has become a framework used for metagenomic data analysis and result visualization. Multivariate analyses such as Linear Discriminate Analysis (LDA) are used to determine which microbial organisms are useful in distinguishing between microbial communities. The differences between communities are visualized in two or three dimensions using dimensional reduction techniques. Other analyses provided by the framework include, but are not limited to, feature selection, cross-validation, multi-objective optimization, side-by-side comparisons of communities, and identification of core members in a microbial community. The effectiveness of these methods and techniques was verified in multiple real world case studies such as body fat classification of elk using a fecal microbiome, identification of important changes in community composition when permafrost is thawed, and longitudinal classification of intestinal locations. The fecal microbiome may be used in the future to assist in assessing the health of animal populations using non-invasive samples. Additionally, the analysis of thawing permafrost may yield insight into the release of greenhouse gases into the atmosphere, furthering our understanding of global warming. Our understanding of the intestinal microbiome may someday grant us understanding and control of our intestinal well being, which plays a significant factor in immune system response and overall health. |
|---|