Wontack Han

Comparative analysis of metagenomes can be used to detect sub-metagenomes(species or genesets) that are associated with specific phenotypes (e.g., host status). The typical workflow is toassemble and annotate metagenomic datasets individually or as a whole, followed by statisticaltests to identify differentially abundant species/genes. The subtractive assembly, a de novo assembly approach for comparative metagenomics, firstly detects differential reads that distinguish between two groups of metagenomes and then only assembles these reads. This approach can make the entire process for finding marker genes/genomes faster and more efficient.

The microbiome research is going through an evolutionary transition from focusing on the characterization of reference microbiomes associated with different environments/hosts to the translational applications, including using microbiome for disease diagnosis, improving the efficacy of cancer treatments, and prevention of diseases (e.g., using probiotics). Microbial markers have been identified from microbiome data derived from cohorts of patients with different diseases, treatment responsiveness, etc, and often predictors based on these markers were built for predicting host phenotype given a microbiome dataset. Unfortunately, these microbial markers and predictors are often not publishedso are not reusable by others. In this project, I report the curation of a repository of microbial marker genes and predictors built from these markers for microbiome-based prediction of host phenotype, and a computational pipeline called Mi2P (from Microbiome to Phenotype) for using the repository.

Due to the complexity of microbial communities, the de novo partition of metagenomic space into specific biological entities remains to be difficult. To address this problem, researchers have utilized various features, including compositional features such as tetra-nucleotide statistics and coverage signals of genetic sequences. However, the assumptions of those methods are not universally true. For example, the methods relying on abundances of genetic sequences are admittedly weak in segregating taxonomically related organisms. In the process of exploring other features, it has been realized that utilizing co-abundance across multiple samples improves the resolution of genome segregation from metagenomic data sets.