Honors College Thesis
 

Mining Public Microbiome Datasets to Identify Specific Microbial Taxa Associated with Anxiety-Related Disorders

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/honors_college_theses/cn69mb516

Descriptions

Attribute NameValues
Creator
Abstract
  • Over the last ten years, clinical, pre-clinical and animal studies have shown associations between the microbiota and neurological functions. Recent work by the scientific community on the gut-microbiome-brain axis have revealed that gut dysbiosis and specific microbial taxa are associated with a myriad of neurological conditions, such as autism spectrum disorder (ASD), anxiety and depression. Many neurological conditions are associated with symptoms of anxiety and stress or are associated with comorbid disorders to anxiety which also correlate with alterations in gut-microbiota composition. In order to study the gut microbiota of an individual, the scientific community widely uses 16S amplicon sequencing of stool samples and comparison of these sequences to a database for taxonomic classification. While 16S amplicon analysis can be limited as it only probes for a small portion of the genomes of microbiota (and subsequently taxonomic classification is usually limited to the genus level), 16S analysis is a widely used means of classifying taxa within the gut microbiota, and a large amount of this data has been made available in public databases. In this project, we leverage available data from multiple studies to determine a common set of 16S amplicons associated with anxiety. To perform this meta-analysis, we used new methods of 16S analysis that have been developed in recent years to extract exact amplicon variants, allowing better comparison of markers across studies. In addition, as many studies in the realm of the gut-brain axis suffer from extremely small samples size, we address these issues with a meta-analysis of 1266 samples containing individuals with anxiety, depression, autism, and ADHD from three studies conducted using a DADA2 pipeline with DESeq2, Metagenomeseq, and ANCOM as a means of differential analysis between individuals with anxiety-related conditions and the neurotypical. Eight different amplicon sequence variants (ASVs) were significant in Metagenomeseq and ANCOM across all datasets. 32 other variants were significant when random subsets of the data were analyzed using similar means. When these both of these ASV groups were used as predictors in a random forest model (using 10 fold cross validation), these ASVs allowed the model to perform better than random at 56% for the eight ASVs found across datasets and greater than 63% when using predictor ASVs found among the random subsets. This study reveals the potential significance of microbial biomarkers of anxiety identified across several studies, identifies significant taxa with the benefits of meta-analyses, and demonstrates the effects these taxa have classification in random forest models.
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Non-Academic Affiliation
Rights Statement
Publisher
Peer Reviewed
Language

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details MartinAustinN2020.pdf 2020-05-28 Public Download